diff --git a/swh/web/browse/utils.py b/swh/web/browse/utils.py index 441f69cd..625db00f 100644 --- a/swh/web/browse/utils.py +++ b/swh/web/browse/utils.py @@ -1,1071 +1,1074 @@ # 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 import magic import math import pypandoc import stat from django.core.cache import cache from django.utils.safestring import mark_safe from importlib import reload 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 ) from swh.web.config import get_config def get_directory_entries(sha1_git): """Function that retrieves the content of a SWH directory from the SWH 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)) entries = sorted(entries, key=lambda e: e['name']) for entry in entries: entry['perms'] = stat.filemode(entry['perms']) dirs = [e for e in entries if e['type'] == 'dir'] files = [e for e in entries if e['type'] == 'file'] 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 as exc: # 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'] def request_content(query_string, max_size=content_display_max_size, raise_if_unavailable=True): """Function that retrieves a SWH content from the SWH 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 as e: pass mimetype = 'unknown' encoding = 'unknown' if filetype: mimetype = filetype['mimetype'] encoding = filetype['encoding'] 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 mimetype.startswith('text/'): # probably a malformed UTF-8 content, reencode 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') 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-swh' 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 SWH 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 SWH 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: 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 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. 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: revision_ids = [] releases_ids = [] snapshot = service.lookup_snapshot(snapshot_id) snapshot_branches = snapshot['branches'] for key in sorted(snapshot_branches.keys()): if not snapshot_branches[key]: continue if snapshot_branches[key]['target_type'] == 'revision': branches.append({'name': key, 'revision': snapshot_branches[key]['target']}) revision_ids.append(snapshot_branches[key]['target']) elif snapshot_branches[key]['target_type'] == 'release': releases_ids.append(snapshot_branches[key]['target']) releases_info = service.lookup_release_multiple(releases_ids) for release in releases_info: releases.append({'name': release['name'], 'date': format_utc_iso_date(release['date']), 'id': release['id'], 'message': release['message'], 'target_type': release['target_type'], 'target': release['target']}) revision_ids.append(release['target']) revisions = service.lookup_revision_multiple(revision_ids) branches_to_remove = [] for idx, revision in enumerate(revisions): if idx < len(branches): if revision: branches[idx]['directory'] = revision['directory'] branches[idx]['date'] = format_utc_iso_date(revision['date']) # noqa branches[idx]['message'] = revision['message'] else: branches_to_remove.append(branches[idx]) else: rel_idx = idx - len(branches) if revision: releases[rel_idx]['directory'] = revision['directory'] for b in branches_to_remove: branches.remove(b) 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. 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 desambiguation 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, link_attrs={}): +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): the text for the produced link + 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 SWH person HTML view to insert in Django templates. Args: person_id (int): a SWH 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', kwargs={'person_id': person_id}, query_params=query_params) return gen_link(person_url, person_name, link_attrs) 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 SWH revision HTML view to insert in Django templates. Args: revision_id (str): a SWH revision id shorten_id (boolean): wheter 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: An HTML link in the form 'revision_id' """ 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']} revision_url = reverse('browse-revision', kwargs={'sha1_git': revision_id}, query_params=query_params) 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 SWH origin HTML view to insert in Django templates. Args: origin_info (dict): a dicted 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', kwargs={'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 SWH 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' """ directory_url = reverse('browse-directory', kwargs={'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 SWH 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', kwargs={'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 SWH 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_type': origin_info['type'], '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', kwargs=url_args, query_params=query_params) else: url_args = {'snapshot_id': snapshot_context['snapshot_id']} directory_url = reverse('browse-snapshot-directory', kwargs=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 SWH 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' """ content_url = reverse('browse-content', kwargs={'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 SWH 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 SWH 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_type': origin_info['type'], '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', kwargs=url_args, query_params=query_params) elif snapshot_context: url_args = {'snapshot_id': snapshot_context['snapshot_id']} revision_log_url = reverse('browse-snapshot-log', kwargs=url_args, query_params=query_params) else: revision_log_url = reverse('browse-revision-log', kwargs={'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 SWH 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' """ 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): revision_log_data = [] for i, log in enumerate(revision_log): if i == per_page: break revision_log_data.append( {'author': gen_person_link(log['author']['id'], log['author']['name'], snapshot_context), 'revision': gen_revision_link(log['id'], True, snapshot_context), 'message': log['message'], 'date': format_utc_iso_date(log['date']), 'directory': log['directory']}) return revision_log_data def prepare_revision_log_for_display(revision_log, per_page, revs_breadcrumb, snapshot_context=None): """ Utility functions that process raw revision log data for HTML display. Its purpose is to: * add links to relevant SWH browse views * format date in human readable format * truncate the message log It also computes the data needed to generate the links for navigating back and forth in the history log. Args: revision_log (list): raw revision log as returned by the SWH web api per_page (int): number of log entries per page revs_breadcrumb (str): breadcrumbs of revisions navigated so far, in the form 'rev1[/rev2/../revN]'. Each revision corresponds to the first one displayed in the HTML view for history log. snapshot_context (dict): if provided, generate snapshot-dependent browsing link """ current_rev = revision_log[0]['id'] next_rev = None prev_rev = None next_revs_breadcrumb = None prev_revs_breadcrumb = None if len(revision_log) == per_page + 1: prev_rev = revision_log[-1]['id'] prev_rev_bc = current_rev if snapshot_context: prev_rev_bc = prev_rev if revs_breadcrumb: revs = revs_breadcrumb.split('/') next_rev = revs[-1] if len(revs) > 1: next_revs_breadcrumb = '/'.join(revs[:-1]) if len(revision_log) == per_page + 1: prev_revs_breadcrumb = revs_breadcrumb + '/' + prev_rev_bc else: prev_revs_breadcrumb = prev_rev_bc return {'revision_log_data': _format_log_entries(revision_log, per_page, snapshot_context), 'prev_rev': prev_rev, 'prev_revs_breadcrumb': prev_revs_breadcrumb, 'next_rev': next_rev, 'next_revs_breadcrumb': next_revs_breadcrumb} # 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'] def get_origin_info(origin_url, origin_type=None): """ Get info about a SWH 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 SWH origin origin_type (str): optionnal origin type Returns: A dict with the following entries: * type: the origin type * url: the origin url * id: the SWH 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 SWH 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 SWH 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 SWH 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 containg url arguments to use when browsing in the context of the origin and its visit """ # 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) origin_info['type'] = origin_type visit_info = get_origin_visit(origin_info, timestamp, visit_id, snapshot_id) visit_info['fmt_date'] = format_utc_iso_date(visit_info['date']) snapshot_id = visit_info['snapshot'] # 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_info['type'], 'origin_url': origin_info['url']} query_params = {'visit_id': visit_id} browse_url = reverse('browse-origin-visits', kwargs=url_args) if timestamp: url_args['timestamp'] = format_utc_iso_date(timestamp, '%Y-%m-%dT%H:%M:%S') visit_url = reverse('browse-origin-directory', kwargs=url_args, query_params=query_params) visit_info['url'] = visit_url branches_url = reverse('browse-origin-branches', kwargs=url_args, query_params=query_params) releases_url = reverse('browse-origin-releases', kwargs=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', kwargs=url_args) branches_url = reverse('browse-snapshot-branches', kwargs=url_args) releases_url = reverse('browse-snapshot-releases', kwargs=url_args) releases = list(reversed(releases)) return { 'swh_type': swh_type, 'snapshot_id': snapshot_id, 'origin_info': origin_info, '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', kwargs={'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', kwargs={'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: 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 = mark_safe('') if swh_object['type'] == 'directory': object_icon = mark_safe('') elif swh_object['type'] == 'release': object_icon = mark_safe('') elif swh_object['type'] == 'revision': object_icon = mark_safe('') # noqa elif swh_object['type'] == 'snapshot': object_icon = mark_safe('') swh_ids.append({ 'object_type': swh_object['type'], 'object_icon': object_icon, 'swh_id': swh_id, 'swh_id_url': reverse('browse-swh-id', kwargs={'swh_id': swh_id}), 'show_options': show_options }) return swh_ids diff --git a/swh/web/browse/views/content.py b/swh/web/browse/views/content.py index 6b0ce9b4..2b59ae24 100644 --- a/swh/web/browse/views/content.py +++ b/swh/web/browse/views/content.py @@ -1,286 +1,293 @@ # 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 difflib import json from distutils.util import strtobool from django.http import HttpResponse from django.utils.safestring import mark_safe from django.shortcuts import render from django.template.defaultfilters import filesizeformat from swh.model.hashutil import hash_to_hex from swh.web.common import query from swh.web.common.utils import ( reverse, gen_path_info ) from swh.web.common.exc import NotFoundExc, handle_view_exception from swh.web.browse.utils import ( request_content, prepare_content_for_display, content_display_max_size, get_snapshot_context, - get_swh_persistent_ids + get_swh_persistent_ids, gen_link ) from swh.web.browse.browseurls import browse_route @browse_route(r'content/(?P.+)/raw/', view_name='browse-content-raw') def content_raw(request, query_string): """Django view that produces a raw display of a SWH content identified by its hash value. The url that points to it is :http:get:`/browse/content/[(algo_hash):](hash)/raw/` """ # noqa try: algo, checksum = query.parse_hash(query_string) checksum = hash_to_hex(checksum) content_data = request_content(query_string, max_size=None) except Exception as exc: return handle_view_exception(request, exc) filename = request.GET.get('filename', None) if not filename: filename = '%s_%s' % (algo, checksum) if content_data['mimetype'].startswith('text/') or \ content_data['mimetype'] == 'inode/x-empty': response = HttpResponse(content_data['raw_data'], content_type="text/plain") response['Content-disposition'] = 'filename=%s' % filename else: response = HttpResponse(content_data['raw_data'], content_type='application/octet-stream') response['Content-disposition'] = 'attachment; filename=%s' % filename return response _auto_diff_size_limit = 20000 @browse_route(r'content/(?P.*)/diff/(?P.*)', # noqa view_name='diff-contents') def _contents_diff(request, from_query_string, to_query_string): """ Browse endpoint used to compute unified diffs between two contents. Diffs are generated only if the two contents are textual. By default, diffs whose size are greater than 20 kB will not be generated. To force the generation of large diffs, the 'force' boolean query parameter must be used. Args: request: input django http request from_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 identifying the first content to_query_string: same as above for identifying the second content Returns: A JSON object containing the unified diff. """ diff_data = {} content_from = None content_to = None content_from_size = 0 content_to_size = 0 content_from_lines = [] content_to_lines = [] force = request.GET.get('force', 'false') path = request.GET.get('path', None) language = 'nohighlight-swh' force = bool(strtobool(force)) if from_query_string == to_query_string: diff_str = 'File renamed without changes' else: try: text_diff = True if from_query_string: content_from = \ request_content(from_query_string, max_size=None) content_from_display_data = \ prepare_content_for_display(content_from['raw_data'], content_from['mimetype'], path) language = content_from_display_data['language'] content_from_size = content_from['length'] if not (content_from['mimetype'].startswith('text/') or content_from['mimetype'] == 'inode/x-empty'): text_diff = False if text_diff and to_query_string: content_to = request_content(to_query_string, max_size=None) content_to_display_data = prepare_content_for_display( content_to['raw_data'], content_to['mimetype'], path) language = content_to_display_data['language'] content_to_size = content_to['length'] if not (content_to['mimetype'].startswith('text/') or content_to['mimetype'] == 'inode/x-empty'): text_diff = False diff_size = abs(content_to_size - content_from_size) if not text_diff: diff_str = 'Diffs are not generated for non textual content' language = 'nohighlight-swh' elif not force and diff_size > _auto_diff_size_limit: diff_str = 'Large diffs are not automatically computed' language = 'nohighlight-swh' else: if content_from: content_from_lines = \ content_from['raw_data'].decode('utf-8')\ .splitlines(True) if content_from_lines and \ content_from_lines[-1][-1] != '\n': content_from_lines[-1] += '[swh-no-nl-marker]\n' if content_to: content_to_lines = content_to['raw_data'].decode('utf-8')\ .splitlines(True) if content_to_lines and content_to_lines[-1][-1] != '\n': content_to_lines[-1] += '[swh-no-nl-marker]\n' diff_lines = difflib.unified_diff(content_from_lines, content_to_lines) diff_str = ''.join(list(diff_lines)[2:]) except Exception as e: diff_str = str(e) diff_data['diff_str'] = diff_str diff_data['language'] = language diff_data_json = json.dumps(diff_data, separators=(',', ': ')) return HttpResponse(diff_data_json, content_type='application/json') @browse_route(r'content/(?P.+)/', view_name='browse-content') def content_display(request, query_string): """Django view that produces an HTML display of a SWH content identified by its hash value. The url that points to it is :http:get:`/browse/content/[(algo_hash):](hash)/` """ # noqa try: algo, checksum = query.parse_hash(query_string) checksum = hash_to_hex(checksum) content_data = request_content(query_string, raise_if_unavailable=False) origin_type = request.GET.get('origin_type', None) origin_url = request.GET.get('origin_url', None) if not origin_url: origin_url = request.GET.get('origin', None) snapshot_context = None if origin_url: try: snapshot_context = get_snapshot_context(None, origin_type, origin_url) except Exception: - raise NotFoundExc('The Software Heritage archive has a content' - ' with the hash you provided but the origin ' - 'mentioned in your request appears broken: ' - '%s. Please check the URL and try again.' % - origin_url) + raw_cnt_url = reverse('browse-content', + kwargs={'query_string': query_string}) + error_message = \ + ('The Software Heritage archive has a content ' + 'with the hash you provided but the origin ' + 'mentioned in your request appears broken: %s. ' + 'Please check the URL and try again.\n\n' + 'Nevertheless, you can still browse the content ' + 'without origin information: %s' + % (gen_link(origin_url), gen_link(raw_cnt_url))) + + raise NotFoundExc(error_message) if snapshot_context: snapshot_context['visit_info'] = None except Exception as exc: return handle_view_exception(request, exc) path = request.GET.get('path', None) content = None language = None mimetype = None if 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'] root_dir = None filename = None path_info = None breadcrumbs = [] if path: split_path = path.split('/') root_dir = split_path[0] filename = split_path[-1] path = path.replace(root_dir + '/', '') path = path[:-len(filename)] path_info = gen_path_info(path) breadcrumbs.append({'name': root_dir[:7], 'url': reverse('browse-directory', kwargs={'sha1_git': root_dir})}) for pi in path_info: breadcrumbs.append({'name': pi['name'], 'url': reverse('browse-directory', kwargs={'sha1_git': root_dir, 'path': pi['path']})}) breadcrumbs.append({'name': filename, 'url': None}) query_params = None if filename: query_params = {'filename': filename} content_raw_url = reverse('browse-content-raw', kwargs={'query_string': query_string}, query_params=query_params) content_metadata = { 'sha1 checksum': content_data['checksums']['sha1'], 'sha1_git checksum': content_data['checksums']['sha1_git'], 'sha256 checksum': content_data['checksums']['sha256'], 'blake2s256 checksum': content_data['checksums']['blake2s256'], 'mime type': content_data['mimetype'], 'encoding': content_data['encoding'], 'size': filesizeformat(content_data['length']), 'language': content_data['language'], 'licenses': content_data['licenses'] } if filename: content_metadata['filename'] = filename sha1_git = content_data['checksums']['sha1_git'] swh_ids = get_swh_persistent_ids([{'type': 'content', 'id': sha1_git}]) heading = 'Content - %s' % sha1_git if breadcrumbs: content_path = '/'.join([bc['name'] for bc in breadcrumbs]) heading += ' - %s' % content_path return render(request, 'content.html', {'heading': heading, 'swh_object_name': 'Content', 'swh_object_metadata': content_metadata, 'content': content, 'content_size': content_data['length'], 'max_content_size': content_display_max_size, 'mimetype': mimetype, 'language': language, 'breadcrumbs': breadcrumbs, 'top_right_link': content_raw_url, 'top_right_link_text': mark_safe( 'Raw File'), 'snapshot_context': snapshot_context, 'vault_cooking': None, 'show_actions_menu': True, 'swh_ids': swh_ids, 'error_code': content_data['error_code'], 'error_message': content_data['error_message'], 'error_description': content_data['error_description']}, status=content_data['error_code']) diff --git a/swh/web/browse/views/directory.py b/swh/web/browse/views/directory.py index 2bef75ad..c2587fa2 100644 --- a/swh/web/browse/views/directory.py +++ b/swh/web/browse/views/directory.py @@ -1,140 +1,148 @@ # 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 from django.shortcuts import render, redirect from django.template.defaultfilters import filesizeformat from swh.web.common import service from swh.web.common.utils import ( reverse, gen_path_info ) -from swh.web.common.exc import handle_view_exception +from swh.web.common.exc import handle_view_exception, NotFoundExc from swh.web.browse.utils import ( get_directory_entries, get_snapshot_context, - get_readme_to_display, get_swh_persistent_ids + get_readme_to_display, get_swh_persistent_ids, + gen_link ) from swh.web.browse.browseurls import browse_route @browse_route(r'directory/(?P[0-9a-f]+)/', r'directory/(?P[0-9a-f]+)/(?P.+)/', view_name='browse-directory') def directory_browse(request, sha1_git, path=None): """Django view for browsing the content of a SWH directory identified by its sha1_git value. The url that points to it is :http:get:`/browse/directory/(sha1_git)/[(path)/]` """ # noqa root_sha1_git = sha1_git try: if path: dir_info = service.lookup_directory_with_path(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', kwargs={'query_string': dir_info['checksums']['sha1']}) return redirect(file_raw_url) sha1_git = dir_info['target'] dirs, files = get_directory_entries(sha1_git) origin_type = request.GET.get('origin_type', None) origin_url = request.GET.get('origin_url', None) if not origin_url: origin_url = request.GET.get('origin', None) snapshot_context = None if origin_url: try: snapshot_context = get_snapshot_context(None, origin_type, origin_url) except Exception: - raise NotFoundExc('The Software Heritage archive has a directory' # noqa - ' with the hash you provided but the origin ' - 'mentioned in your request appears broken: ' - '%s. Please check the URL and try again.' % - origin_url) + raw_dir_url = reverse('browse-directory', + kwargs={'sha1_git': sha1_git}) + error_message = \ + ('The Software Heritage archive has a directory ' + 'with the hash you provided but the origin ' + 'mentioned in your request appears broken: %s. ' + 'Please check the URL and try again.\n\n' + 'Nevertheless, you can still browse the directory ' + 'without origin information: %s' + % (gen_link(origin_url), gen_link(raw_dir_url))) + + raise NotFoundExc(error_message) if snapshot_context: snapshot_context['visit_info'] = None except Exception as exc: return handle_view_exception(request, exc) path_info = gen_path_info(path) breadcrumbs = [] breadcrumbs.append({'name': root_sha1_git[:7], 'url': reverse('browse-directory', kwargs={'sha1_git': root_sha1_git})}) for pi in path_info: breadcrumbs.append({'name': pi['name'], 'url': reverse('browse-directory', kwargs={'sha1_git': root_sha1_git, 'path': pi['path']})}) path = '' if path is None else (path + '/') for d in dirs: d['url'] = reverse('browse-directory', kwargs={'sha1_git': root_sha1_git, 'path': path + d['name']}) sum_file_sizes = 0 readmes = {} for f in files: query_string = 'sha1_git:' + f['target'] f['url'] = reverse('browse-content', kwargs={'query_string': query_string}, query_params={'path': root_sha1_git + '/' + path + f['name']}) 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) sum_file_sizes = filesizeformat(sum_file_sizes) dir_metadata = {'id': sha1_git, 'number of regular files': len(files), 'number of subdirectories': len(dirs), 'sum of regular file sizes': sum_file_sizes} vault_cooking = { 'directory_context': True, 'directory_id': sha1_git, 'revision_context': False, 'revision_id': None } swh_ids = get_swh_persistent_ids([{'type': 'directory', 'id': sha1_git}]) heading = 'Directory - %s' % sha1_git if breadcrumbs: dir_path = '/'.join([bc['name'] for bc in breadcrumbs]) + '/' heading += ' - %s' % dir_path return render(request, 'directory.html', {'heading': heading, 'swh_object_name': 'Directory', 'swh_object_metadata': dir_metadata, 'dirs': dirs, 'files': files, 'breadcrumbs': breadcrumbs, 'top_right_link': None, 'top_right_link_text': None, '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}) diff --git a/swh/web/browse/views/release.py b/swh/web/browse/views/release.py index b8842a68..44dc2b6f 100644 --- a/swh/web/browse/views/release.py +++ b/swh/web/browse/views/release.py @@ -1,171 +1,178 @@ # 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 from django.shortcuts import render from django.utils.safestring import mark_safe from swh.web.common import service from swh.web.common.utils import ( reverse, format_utc_iso_date ) from swh.web.common.exc import NotFoundExc, handle_view_exception from swh.web.browse.browseurls import browse_route from swh.web.browse.utils import ( gen_person_link, gen_revision_link, get_snapshot_context, gen_link, gen_snapshot_link, get_swh_persistent_ids ) @browse_route(r'release/(?P[0-9a-f]+)/', view_name='browse-release') def release_browse(request, sha1_git): """ Django view that produces an HTML display of a SWH release identified by its id. The url that points to it is :http:get:`/browse/release/(sha1_git)/`. """ try: release = service.lookup_release(sha1_git) snapshot_context = None origin_info = None snapshot_id = request.GET.get('snapshot_id', None) origin_type = request.GET.get('origin_type', None) origin_url = request.GET.get('origin_url', None) if not origin_url: origin_url = request.GET.get('origin', None) timestamp = request.GET.get('timestamp', None) visit_id = request.GET.get('visit_id', None) if origin_url: try: snapshot_context = get_snapshot_context(snapshot_id, origin_type, # noqa origin_url, timestamp, visit_id) except Exception: - raise NotFoundExc('The Software Heritage archive has a release' - ' with the hash you provided but the origin ' - 'mentioned in your request appears broken: ' - '%s. Please check the URL and try again.' % - origin_url) + raw_rel_url = reverse('browse-release', + kwargs={'sha1_git': sha1_git}) + error_message = \ + ('The Software Heritage archive has a release ' + 'with the hash you provided but the origin ' + 'mentioned in your request appears broken: %s. ' + 'Please check the URL and try again.\n\n' + 'Nevertheless, you can still browse the release ' + 'without origin information: %s' + % (gen_link(origin_url), gen_link(raw_rel_url))) + + raise NotFoundExc(error_message) origin_info = snapshot_context['origin_info'] elif snapshot_id: snapshot_context = get_snapshot_context(snapshot_id) except Exception as exc: return handle_view_exception(request, exc) release_data = {} release_data['author'] = \ gen_person_link(release['author']['id'], release['author']['name'], snapshot_context) release_data['date'] = format_utc_iso_date(release['date']) release_data['id'] = sha1_git release_data['name'] = release['name'] release_data['synthetic'] = release['synthetic'] release_data['target type'] = release['target_type'] if release['target_type'] == 'revision': release_data['target'] = \ gen_revision_link(release['target'], snapshot_context=snapshot_context) elif release['target_type'] == 'content': content_url = \ reverse('browse-content', kwargs={'query_string': 'sha1_git:' + release['target']}) release_data['target'] = gen_link(content_url, release['target']) elif release['target_type'] == 'directory': directory_url = \ reverse('browse-directory', kwargs={'sha1_git': release['target']}) release_data['target'] = gen_link(directory_url, release['target']) elif release['target_type'] == 'release': release_url = \ reverse('browse-release', kwargs={'sha1_git': release['target']}) release_data['target'] = gen_link(release_url, release['target']) release_note_lines = [] if release['message']: release_note_lines = release['message'].split('\n') vault_cooking = None release_target_link = 'Target: ' if release['target_type'] == 'revision': release_target_link += '' # noqa try: revision = service.lookup_revision(release['target']) vault_cooking = { 'directory_context': True, 'directory_id': revision['directory'], 'revision_context': True, 'revision_id': release['target'] } except Exception: pass else: release_target_link += release['target_type'] release_target_link += ' ' + release_data['target'] if snapshot_context: release_data['snapshot id'] = snapshot_context['snapshot_id'] if origin_info: release_url = reverse('browse-release', kwargs={'sha1_git': release['id']}) release_data['context-independent release'] = \ gen_link(release_url, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}) release_data['origin id'] = origin_info['id'] release_data['origin type'] = origin_info['type'] release_data['origin url'] = gen_link(origin_info['url'], origin_info['url']) browse_snapshot_link = \ gen_snapshot_link(snapshot_context['snapshot_id'], link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}) release_data['snapshot'] = browse_snapshot_link swh_objects = [{'type': 'release', 'id': sha1_git}] if snapshot_context: snapshot_id = snapshot_context['snapshot_id'] if snapshot_id: swh_objects.append({'type': 'snapshot', 'id': snapshot_id}) swh_ids = get_swh_persistent_ids(swh_objects, snapshot_context) release_note_header = 'None' if len(release_note_lines) > 0: release_note_header = release_note_lines[0] heading = 'Release - %s' % release['name'] if snapshot_context: context_found = 'snapshot: %s' % snapshot_context['snapshot_id'] if origin_info: context_found = 'origin: %s' % origin_info['url'] heading += ' - %s' % context_found return render(request, 'release.html', {'heading': heading, 'swh_object_name': 'Release', 'swh_object_metadata': release_data, 'release_name': release['name'], 'release_note_header': release_note_header, 'release_note_body': '\n'.join(release_note_lines[1:]), 'release_target_link': mark_safe(release_target_link), 'snapshot_context': snapshot_context, 'show_actions_menu': True, 'breadcrumbs': None, 'vault_cooking': vault_cooking, 'top_right_link': None, 'top_right_link_text': None, 'swh_ids': swh_ids}) diff --git a/swh/web/browse/views/revision.py b/swh/web/browse/views/revision.py index 2b73e29b..4d1a1930 100644 --- a/swh/web/browse/views/revision.py +++ b/swh/web/browse/views/revision.py @@ -1,523 +1,529 @@ # 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 hashlib import json import textwrap from django.http import HttpResponse from django.shortcuts import render, redirect from django.template.defaultfilters import filesizeformat from django.utils.safestring import mark_safe from swh.web.common import service from swh.web.common.utils import ( reverse, format_utc_iso_date, gen_path_info ) from swh.web.common.exc import NotFoundExc, handle_view_exception from swh.web.browse.browseurls import browse_route from swh.web.browse.utils import ( gen_link, gen_person_link, gen_revision_link, prepare_revision_log_for_display, get_snapshot_context, gen_snapshot_directory_link, get_revision_log_url, get_directory_entries, gen_directory_link, request_content, prepare_content_for_display, content_display_max_size, gen_snapshot_link, get_readme_to_display, get_swh_persistent_ids ) def _gen_content_url(revision, query_string, path, snapshot_context): if snapshot_context: url_args = snapshot_context['url_args'] url_args['path'] = path query_params = snapshot_context['query_params'] query_params['revision'] = revision['id'] content_url = reverse('browse-origin-content', kwargs=url_args, query_params=query_params) else: content_path = '%s/%s' % (revision['directory'], path) content_url = reverse('browse-content', kwargs={'query_string': query_string}, query_params={'path': content_path}) return content_url def _gen_diff_link(idx, diff_anchor, link_text): if idx < _max_displayed_file_diffs: return gen_link(diff_anchor, link_text) else: return link_text # TODO: put in conf _max_displayed_file_diffs = 1000 def _gen_revision_changes_list(revision, changes, snapshot_context): """ Returns a HTML string describing the file changes introduced in a revision. As this string will be displayed in the browse revision view, links to adequate file diffs are also generated. Args: revision (str): hexadecimal representation of a revision identifier changes (list): list of file changes in the revision snapshot_context (dict): optional origin context used to reverse the content urls Returns: A string to insert in a revision HTML view. """ changes_msg = [] for i, change in enumerate(changes): hasher = hashlib.sha1() from_query_string = '' to_query_string = '' diff_id = 'diff-' if change['from']: from_query_string = 'sha1_git:' + change['from']['target'] diff_id += change['from']['target'] + '-' + change['from_path'] diff_id += '-' if change['to']: to_query_string = 'sha1_git:' + change['to']['target'] diff_id += change['to']['target'] + change['to_path'] change['path'] = change['to_path'] or change['from_path'] url_args = {'from_query_string': from_query_string, 'to_query_string': to_query_string} query_params = {'path': change['path']} change['diff_url'] = reverse('diff-contents', kwargs=url_args, query_params=query_params) hasher.update(diff_id.encode('utf-8')) diff_id = hasher.hexdigest() change['id'] = diff_id panel_diff_link = '#panel_' + diff_id if change['type'] == 'modify': change['content_url'] = \ _gen_content_url(revision, to_query_string, change['to_path'], snapshot_context) changes_msg.append('modified: %s' % _gen_diff_link(i, panel_diff_link, change['to_path'])) elif change['type'] == 'insert': change['content_url'] = \ _gen_content_url(revision, to_query_string, change['to_path'], snapshot_context) changes_msg.append('new file: %s' % _gen_diff_link(i, panel_diff_link, change['to_path'])) elif change['type'] == 'delete': parent = service.lookup_revision(revision['parents'][0]) change['content_url'] = \ _gen_content_url(parent, from_query_string, change['from_path'], snapshot_context) changes_msg.append('deleted: %s' % _gen_diff_link(i, panel_diff_link, change['from_path'])) elif change['type'] == 'rename': change['content_url'] = \ _gen_content_url(revision, to_query_string, change['to_path'], snapshot_context) link_text = change['from_path'] + ' → ' + change['to_path'] changes_msg.append('renamed: %s' % _gen_diff_link(i, panel_diff_link, link_text)) if not changes: changes_msg.append('No changes') return mark_safe('\n'.join(changes_msg)) @browse_route(r'revision/(?P[0-9a-f]+)/diff/', view_name='diff-revision') def _revision_diff(request, sha1_git): """ Browse internal endpoint to compute revision diff """ try: revision = service.lookup_revision(sha1_git) snapshot_context = None origin_type = request.GET.get('origin_type', None) origin_url = request.GET.get('origin_url', None) if not origin_url: origin_url = request.GET.get('origin', None) timestamp = request.GET.get('timestamp', None) visit_id = request.GET.get('visit_id', None) if origin_url: snapshot_context = get_snapshot_context(None, origin_type, origin_url, timestamp, visit_id) except Exception as exc: return handle_view_exception(request, exc) changes = service.diff_revision(sha1_git) changes_msg = _gen_revision_changes_list(revision, changes, snapshot_context) diff_data = { 'total_nb_changes': len(changes), 'changes': changes[:_max_displayed_file_diffs], 'changes_msg': changes_msg } diff_data_json = json.dumps(diff_data, separators=(',', ': ')) return HttpResponse(diff_data_json, content_type='application/json') NB_LOG_ENTRIES = 20 @browse_route(r'revision/(?P[0-9a-f]+)/log/', view_name='browse-revision-log') def revision_log_browse(request, sha1_git): """ Django view that produces an HTML display of the history log for a SWH revision identified by its id. The url that points to it is :http:get:`/browse/revision/(sha1_git)/log/`. """ # noqa try: per_page = int(request.GET.get('per_page', NB_LOG_ENTRIES)) revision_log = service.lookup_revision_log(sha1_git, limit=per_page+1) revision_log = list(revision_log) except Exception as exc: return handle_view_exception(request, exc) revs_breadcrumb = request.GET.get('revs_breadcrumb', None) revision_log_display_data = prepare_revision_log_for_display( revision_log, per_page, revs_breadcrumb) prev_rev = revision_log_display_data['prev_rev'] prev_revs_breadcrumb = revision_log_display_data['prev_revs_breadcrumb'] prev_log_url = None if prev_rev: prev_log_url = \ reverse('browse-revision-log', kwargs={'sha1_git': prev_rev}, query_params={'revs_breadcrumb': prev_revs_breadcrumb, 'per_page': per_page}) next_rev = revision_log_display_data['next_rev'] next_revs_breadcrumb = revision_log_display_data['next_revs_breadcrumb'] next_log_url = None if next_rev: next_log_url = \ reverse('browse-revision-log', kwargs={'sha1_git': next_rev}, query_params={'revs_breadcrumb': next_revs_breadcrumb, 'per_page': per_page}) revision_log_data = revision_log_display_data['revision_log_data'] for log in revision_log_data: log['directory'] = gen_directory_link( log['directory'], link_text='Browse files', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}) return render(request, 'revision-log.html', {'heading': 'Revision history', 'swh_object_name': 'Revision history', 'swh_object_metadata': None, 'revision_log': revision_log_data, 'next_log_url': next_log_url, 'prev_log_url': prev_log_url, 'breadcrumbs': None, 'top_right_link': None, 'top_right_link_text': None, 'snapshot_context': None, 'vault_cooking': None, 'show_actions_menu': True, 'swh_ids': None}) @browse_route(r'revision/(?P[0-9a-f]+)/', r'revision/(?P[0-9a-f]+)/(?P.+)/', view_name='browse-revision') def revision_browse(request, sha1_git, extra_path=None): """ Django view that produces an HTML display of a SWH revision identified by its id. The url that points to it is :http:get:`/browse/revision/(sha1_git)/`. """ try: revision = service.lookup_revision(sha1_git) # some readme files can reference assets reachable from the # browsed directory, handle that special case in order to # correctly displayed them if extra_path: dir_info = \ service.lookup_directory_with_path(revision['directory'], extra_path) if dir_info and dir_info['type'] == 'file': file_raw_url = reverse( 'browse-content-raw', kwargs={'query_string': dir_info['checksums']['sha1']}) return redirect(file_raw_url) origin_info = None snapshot_context = None origin_type = request.GET.get('origin_type', None) origin_url = request.GET.get('origin_url', None) if not origin_url: origin_url = request.GET.get('origin', None) timestamp = request.GET.get('timestamp', None) visit_id = request.GET.get('visit_id', None) snapshot_id = request.GET.get('snapshot_id', None) path = request.GET.get('path', None) dir_id = None dirs, files = None, None content_data = None if origin_url: try: snapshot_context = get_snapshot_context(None, origin_type, origin_url, timestamp, visit_id) except Exception: - raise NotFoundExc('The Software Heritage archive has a revision' # noqa - ' with the hash you provided but the origin ' - 'mentioned in your request appears broken: ' - '%s. Please check the URL and try again.' % - origin_url) + raw_rev_url = reverse('browse-revision', + kwargs={'sha1_git': sha1_git}) + error_message = \ + ('The Software Heritage archive has a revision ' + 'with the hash you provided but the origin ' + 'mentioned in your request appears broken: %s. ' + 'Please check the URL and try again.\n\n' + 'Nevertheless, you can still browse the revision ' + 'without origin information: %s' + % (gen_link(origin_url), gen_link(raw_rev_url))) + raise NotFoundExc(error_message) origin_info = snapshot_context['origin_info'] snapshot_id = snapshot_context['snapshot_id'] elif snapshot_id: snapshot_context = get_snapshot_context(snapshot_id) if path: file_info = \ service.lookup_directory_with_path(revision['directory'], path) if file_info['type'] == 'dir': dir_id = file_info['target'] else: query_string = 'sha1_git:' + file_info['target'] content_data = request_content(query_string, raise_if_unavailable=False) else: dir_id = revision['directory'] if dir_id: path = '' if path is None else (path + '/') dirs, files = get_directory_entries(dir_id) except Exception as exc: return handle_view_exception(request, exc) revision_data = {} revision_data['author'] = \ gen_person_link(revision['author']['id'], revision['author']['name'], snapshot_context) revision_data['committer'] = \ gen_person_link(revision['committer']['id'], revision['committer']['name'], snapshot_context) revision_data['committer date'] = format_utc_iso_date( revision['committer_date']) revision_data['date'] = format_utc_iso_date(revision['date']) if snapshot_context: revision_data['snapshot id'] = snapshot_id revision_data['directory'] = \ gen_snapshot_directory_link(snapshot_context, sha1_git, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', # noqa 'role': 'button'}) else: revision_data['directory'] = \ gen_directory_link(revision['directory'], link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}) revision_data['id'] = sha1_git revision_data['merge'] = revision['merge'] revision_data['metadata'] = json.dumps(revision['metadata'], sort_keys=True, indent=4, separators=(',', ': ')) if origin_info: revision_data['context-independent revision'] = \ gen_revision_link(sha1_git, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}) revision_data['origin id'] = origin_info['id'] revision_data['origin type'] = origin_info['type'] revision_data['origin url'] = gen_link(origin_info['url'], origin_info['url']) browse_snapshot_link = \ gen_snapshot_link(snapshot_id, link_text='Browse', link_attrs={'class': 'btn btn-default btn-sm', 'role': 'button'}) revision_data['snapshot'] = browse_snapshot_link parents = '' for p in revision['parents']: parent_link = gen_revision_link(p, snapshot_context=snapshot_context) parents += parent_link + '
' revision_data['parents'] = mark_safe(parents) revision_data['synthetic'] = revision['synthetic'] revision_data['type'] = revision['type'] message_lines = revision['message'].split('\n') parents_links = '%s parent%s ' % \ (len(revision['parents']), '' if len(revision['parents']) == 1 else 's') parents_links += ' ' for p in revision['parents']: parent_link = gen_revision_link(p, shorten_id=True, snapshot_context=snapshot_context) parents_links += parent_link if p != revision['parents'][-1]: parents_links += ' + ' path_info = gen_path_info(path) query_params = {'snapshot_id': snapshot_id, 'origin_type': origin_type, 'origin': origin_url, 'timestamp': timestamp, 'visit_id': visit_id} breadcrumbs = [] breadcrumbs.append({'name': revision['directory'][:7], 'url': reverse('browse-revision', kwargs={'sha1_git': sha1_git}, query_params=query_params)}) for pi in path_info: query_params['path'] = pi['path'] breadcrumbs.append({'name': pi['name'], 'url': reverse('browse-revision', kwargs={'sha1_git': sha1_git}, query_params=query_params)}) vault_cooking = { 'directory_context': False, 'directory_id': None, 'revision_context': True, 'revision_id': sha1_git } swh_objects = [{'type': 'revision', 'id': sha1_git}] content = None content_size = None mimetype = None language = None readme_name = None readme_url = None readme_html = None readmes = {} error_code = 200 error_message = '' error_description = '' if content_data: breadcrumbs[-1]['url'] = None content_size = content_data['length'] mimetype = content_data['mimetype'] if content_data['raw_data']: 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'] query_params = {} if path: query_params['filename'] = path_info[-1]['name'] top_right_link = reverse('browse-content-raw', kwargs={'query_string': query_string}, query_params=query_params) top_right_link_text = mark_safe( 'Raw File') swh_objects.append({'type': 'content', 'id': file_info['target']}) error_code = content_data['error_code'] error_message = content_data['error_message'] error_description = content_data['error_description'] else: for d in dirs: query_params['path'] = path + d['name'] d['url'] = reverse('browse-revision', kwargs={'sha1_git': sha1_git}, query_params=query_params) for f in files: query_params['path'] = path + f['name'] f['url'] = reverse('browse-revision', kwargs={'sha1_git': sha1_git}, query_params=query_params) 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) top_right_link = get_revision_log_url(sha1_git, snapshot_context) top_right_link_text = mark_safe( '' 'History') vault_cooking['directory_context'] = True vault_cooking['directory_id'] = dir_id swh_objects.append({'type': 'directory', 'id': dir_id}) diff_revision_url = reverse('diff-revision', kwargs={'sha1_git': sha1_git}, query_params={'origin_type': origin_type, 'origin': origin_url, 'timestamp': timestamp, 'visit_id': visit_id}) if snapshot_id: swh_objects.append({'type': 'snapshot', 'id': snapshot_id}) swh_ids = get_swh_persistent_ids(swh_objects, snapshot_context) heading = 'Revision - %s - %s' %\ (sha1_git[:7], textwrap.shorten(message_lines[0], width=70)) if snapshot_context: context_found = 'snapshot: %s' % snapshot_context['snapshot_id'] if origin_info: context_found = 'origin: %s' % origin_info['url'] heading += ' - %s' % context_found return render(request, 'revision.html', {'heading': heading, 'swh_object_name': 'Revision', 'swh_object_metadata': revision_data, 'message_header': message_lines[0], 'message_body': '\n'.join(message_lines[1:]), 'parents_links': mark_safe(parents_links), 'snapshot_context': snapshot_context, 'dirs': dirs, 'files': files, 'content': content, 'content_size': content_size, 'max_content_size': content_display_max_size, 'mimetype': mimetype, 'language': language, 'readme_name': readme_name, 'readme_url': readme_url, 'readme_html': readme_html, 'breadcrumbs': breadcrumbs, 'top_right_link': top_right_link, 'top_right_link_text': top_right_link_text, 'vault_cooking': vault_cooking, 'diff_revision_url': diff_revision_url, 'show_actions_menu': True, 'swh_ids': swh_ids, 'error_code': error_code, 'error_message': error_message, 'error_description': error_description}, status=error_code) diff --git a/swh/web/common/exc.py b/swh/web/common/exc.py index 481a03e6..0696fe03 100644 --- a/swh/web/common/exc.py +++ b/swh/web/common/exc.py @@ -1,111 +1,113 @@ # 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 traceback from django.shortcuts import render +from django.utils.safestring import mark_safe from swh.web.config import get_config class BadInputExc(ValueError): """Wrong request to the api. Example: Asking a content with the wrong identifier format. """ pass class NotFoundExc(Exception): """Good request to the api but no result were found. Example: Asking a content with the right identifier format but that content does not exist. """ pass class ForbiddenExc(Exception): """Good request to the api, forbidden result to return due to enforce policy. Example: Asking for a raw content which exists but whose mimetype is not text. """ pass http_status_code_message = { 400: "Bad Request", 401: "Unauthorized", 403: "Access Denied", 404: "Resource not found", 500: "Internal Server Error", 501: "Not Implemented", 502: "Bad Gateway", 503: "Service unavailable" } def _generate_error_page(request, error_code, error_description): return render(request, "error.html", {'error_code': error_code, 'error_message': http_status_code_message[error_code], - 'error_description': error_description}, status=error_code) + 'error_description': mark_safe(error_description)}, + status=error_code) def swh_handle400(request): """ Custom Django HTTP error 400 handler for swh-web. """ error_description = 'The server cannot process the request to %s due to '\ 'something that is perceived to be a client error.' %\ request.META['PATH_INFO'] return _generate_error_page(request, 400, error_description) def swh_handle403(request): """ Custom Django HTTP error 403 handler for swh-web. """ error_description = 'The resource %s requires an authentication.' %\ request.META['PATH_INFO'] return _generate_error_page(request, 403, error_description) def swh_handle404(request): """ Custom Django HTTP error 404 handler for swh-web. """ error_description = 'The resource %s could not be found on the server.' %\ request.META['PATH_INFO'] return _generate_error_page(request, 404, error_description) def swh_handle500(request): """ Custom Django HTTP error 500 handler for swh-web. """ error_description = 'An unexpected condition was encountered when '\ 'requesting resource %s.' %\ request.META['PATH_INFO'] return _generate_error_page(request, 500, error_description) def handle_view_exception(request, exc): """ Function used to generate an error page when an exception was raised inside a swh-web browse view. """ error_code = 400 error_description = str(exc) if get_config()['debug']: error_description = traceback.format_exc() if isinstance(exc, NotFoundExc): error_code = 404 return _generate_error_page(request, error_code, error_description)