diff --git a/swh/web/browse/snapshot_context.py b/swh/web/browse/snapshot_context.py index 148b29bc..58116705 100644 --- a/swh/web/browse/snapshot_context.py +++ b/swh/web/browse/snapshot_context.py @@ -1,1330 +1,1334 @@ -# Copyright (C) 2018-2021 The Software Heritage developers +# Copyright (C) 2018-2022 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 for browsing the archive in a snapshot context. from collections import defaultdict from typing import Any, Dict, List, Optional, Tuple from django.core.cache import cache from django.shortcuts import render from django.utils.html import escape from swh.model.hashutil import hash_to_bytes from swh.model.model import Snapshot from swh.model.swhids import CoreSWHID, ObjectType from swh.web.browse.utils import ( format_log_entries, gen_release_link, gen_revision_link, gen_revision_log_link, gen_revision_url, gen_snapshot_link, get_directory_entries, get_readme_to_display, ) from swh.web.common import archive from swh.web.common.exc import BadInputExc, NotFoundExc, http_status_code_message from swh.web.common.identifiers import get_swhids_info from swh.web.common.origin_visits import get_origin_visit from swh.web.common.typing import ( DirectoryMetadata, OriginInfo, SnapshotBranchInfo, SnapshotContext, SnapshotReleaseInfo, SWHObjectInfo, ) from swh.web.common.utils import ( format_utc_iso_date, gen_path_info, reverse, swh_object_icons, ) from swh.web.config import get_config _empty_snapshot_id = Snapshot(branches={}).id.hex() def _get_branch(branches, branch_name, snapshot_id): """ Utility function to get a specific branch from a snapshot. Returns None if the branch cannot be found. """ filtered_branches = [b for b in branches if b["name"] == branch_name] if filtered_branches: return filtered_branches[0] else: # case where a large branches list has been truncated snp = archive.lookup_snapshot( snapshot_id, branches_from=branch_name, branches_count=1, target_types=["revision", "alias"], # pull request branches must be browsable even if they are hidden # by default in branches list branch_name_exclude_prefix=None, ) snp_branch, _, _ = process_snapshot_branches(snp) if snp_branch and snp_branch[0]["name"] == branch_name: branches.append(snp_branch[0]) return snp_branch[0] def _get_release(releases, release_name, snapshot_id): """ Utility function to get a specific release from a snapshot. Returns None if the release cannot be found. """ filtered_releases = [r for r in releases if r["name"] == release_name] if filtered_releases: return filtered_releases[0] else: # case where a large branches list has been truncated try: # git origins have specific branches for releases snp = archive.lookup_snapshot( snapshot_id, branches_from=f"refs/tags/{release_name}", branches_count=1, target_types=["release"], ) except NotFoundExc: snp = archive.lookup_snapshot( snapshot_id, branches_from=release_name, branches_count=1, target_types=["release", "alias"], ) _, snp_release, _ = process_snapshot_branches(snp) if snp_release and snp_release[0]["name"] == release_name: releases.append(snp_release[0]) return snp_release[0] def _branch_not_found( branch_type, branch, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id ): """ 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" target_type = "revision" else: branch_type = "Release" branch_type_plural = "releases" target_type = "release" if snapshot_id and snapshot_sizes[target_type] == 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 snapshot_sizes[target_type] == 0: msg = ( "Origin with url %s" " for visit with id %s has an empty list" " of %s!" % (origin_info["url"], visit_id, branch_type_plural) ) elif visit_id: msg = ( "%s %s associated to visit with" " id %s for origin with url %s" " not found!" % (branch_type, branch, visit_id, origin_info["url"]) ) elif snapshot_sizes[target_type] == 0: msg = ( "Origin with url %s" " for visit with timestamp %s has an empty list" " of %s!" % (origin_info["url"], timestamp, branch_type_plural) ) else: msg = ( "%s %s associated to visit with" " timestamp %s for origin with " "url %s not found!" % (branch_type, branch, timestamp, origin_info["url"]) ) raise NotFoundExc(escape(msg)) def process_snapshot_branches( snapshot: Dict[str, Any] ) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo], Dict[str, Any]]: """ 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: A dict describing a snapshot as returned for instance by :func:`swh.web.common.archive.lookup_snapshot` Returns: A tuple whose first member is the sorted list of branches targeting revisions, second member the sorted list of branches targeting releases and third member a dict mapping resolved branch aliases to their real target. """ snapshot_branches = snapshot["branches"] branches: Dict[str, SnapshotBranchInfo] = {} branch_aliases: Dict[str, str] = {} releases: Dict[str, SnapshotReleaseInfo] = {} 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] = SnapshotBranchInfo( name=branch_name, alias=False, revision=target_id, date=None, directory=None, message=None, url=None, ) revision_to_branch[target_id].add(branch_name) elif target_type == "release": release_to_branch[target_id].add(branch_name) elif target_type == "alias": branch_aliases[branch_name] = target_id # FIXME: handle pointers to other object types def _add_release_info(branch, release, alias=False): releases[branch] = SnapshotReleaseInfo( name=release["name"], alias=alias, branch_name=branch, date=format_utc_iso_date(release["date"]), directory=None, id=release["id"], message=release["message"], target_type=release["target_type"], target=release["target"], url=None, ) def _add_branch_info(branch, revision, alias=False): branches[branch] = SnapshotBranchInfo( name=branch, alias=alias, revision=revision["id"], directory=revision["directory"], date=format_utc_iso_date(revision["date"]), message=revision["message"], url=None, ) releases_info = archive.lookup_release_multiple(release_to_branch.keys()) for release in releases_info: if release is None: continue branches_to_update = release_to_branch[release["id"]] for branch in branches_to_update: _add_release_info(branch, release) if release["target_type"] == "revision": revision_to_release[release["target"]].update(branches_to_update) revisions = archive.lookup_revision_multiple( set(revision_to_branch.keys()) | set(revision_to_release.keys()) ) for revision in revisions: if not revision: continue for branch in revision_to_branch[revision["id"]]: _add_branch_info(branch, revision) for release_id in revision_to_release[revision["id"]]: releases[release_id]["directory"] = revision["directory"] resolved_aliases = {} for branch_alias, branch_target in branch_aliases.items(): resolved_alias = archive.lookup_snapshot_alias(snapshot["id"], branch_alias) resolved_aliases[branch_alias] = resolved_alias if resolved_alias is None: continue target_type = resolved_alias["target_type"] target = resolved_alias["target"] if target_type == "revision": revision = archive.lookup_revision(target) _add_branch_info(branch_alias, revision, alias=True) elif target_type == "release": release = archive.lookup_release(target) _add_release_info(branch_alias, release, alias=True) 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, resolved_aliases def get_snapshot_content( snapshot_id: str, ) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo], Dict[str, Any]]: """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: hexadecimal representation of the snapshot identifier Returns: A tuple with three 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. The third one is a dict mapping resolved branch aliases to their real target. 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"], cache_entry.get("aliases", {}), ) branches: List[SnapshotBranchInfo] = [] releases: List[SnapshotReleaseInfo] = [] aliases: Dict[str, Any] = {} snapshot_content_max_size = get_config()["snapshot_content_max_size"] if snapshot_id: snapshot = archive.lookup_snapshot( snapshot_id, branches_count=snapshot_content_max_size ) branches, releases, aliases = process_snapshot_branches(snapshot) cache.set( cache_entry_id, {"branches": branches, "releases": releases, "aliases": aliases} ) return branches, releases, aliases def get_origin_visit_snapshot( origin_info: OriginInfo, visit_ts: Optional[str] = None, visit_id: Optional[int] = None, snapshot_id: Optional[str] = None, ) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo], Dict[str, Any]]: """Returns the lists of branches and releases associated to an origin for a given visit. The visit is expressed by either: * a snapshot identifier * a timestamp, if no visit with that exact timestamp is found, the closest one 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: a dict filled with origin information visit_ts: an ISO 8601 datetime string to parse visit_id: visit id for disambiguation in case several visits have the same timestamp snapshot_id: if provided, visit associated to the snapshot will be processed Returns: A tuple with three 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. The third one is a dict mapping resolved branch aliases to their real target. 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 get_snapshot_context( snapshot_id: Optional[str] = None, origin_url: Optional[str] = None, timestamp: Optional[str] = None, visit_id: Optional[int] = None, branch_name: Optional[str] = None, release_name: Optional[str] = None, revision_id: Optional[str] = None, path: Optional[str] = None, browse_context: str = "directory", ) -> SnapshotContext: """ 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: hexadecimal representation of a snapshot identifier origin_url: an origin_url timestamp: a datetime string for retrieving the closest visit of the origin visit_id: optional visit id for disambiguation in case of several visits with the same timestamp branch_name: optional branch name set when browsing the snapshot in that scope (will default to "HEAD" if not provided) release_name: optional release name set when browsing the snapshot in that scope revision_id: optional revision identifier set when browsing the snapshot in that scope path: optional path of the object currently browsed in the snapshot browse_context: indicates which type of object is currently browsed Returns: A dict filled with snapshot context information. Raises: swh.web.common.exc.NotFoundExc: if no snapshot is found for the visit of an origin. """ assert origin_url is not None or snapshot_id is not None origin_info = None visit_info = None url_args = {} query_params: Dict[str, Any] = {} origin_visits_url = None if origin_url: if visit_id is not None: query_params["visit_id"] = visit_id elif snapshot_id is not None: query_params["snapshot"] = snapshot_id origin_info = archive.lookup_origin({"url": origin_url}) visit_info = get_origin_visit(origin_info, timestamp, visit_id, snapshot_id) formatted_date = format_utc_iso_date(visit_info["date"]) visit_info["formatted_date"] = formatted_date snapshot_id = visit_info["snapshot"] if not snapshot_id: raise NotFoundExc( "No snapshot associated to the visit of origin " "%s on %s" % (escape(origin_url), formatted_date) ) # provided timestamp is not necessarily equals to the one # of the retrieved visit, so get the exact one in order # to use it in the urls generated below if timestamp: timestamp = visit_info["date"] branches, releases, aliases = get_origin_visit_snapshot( origin_info, timestamp, visit_id, snapshot_id ) query_params["origin_url"] = origin_info["url"] origin_visits_url = reverse( "browse-origin-visits", query_params={"origin_url": origin_info["url"]} ) if timestamp is not None: query_params["timestamp"] = format_utc_iso_date( timestamp, "%Y-%m-%dT%H:%M:%SZ" ) visit_url = reverse("browse-origin-directory", query_params=query_params) visit_info["url"] = directory_url = visit_url branches_url = reverse("browse-origin-branches", query_params=query_params) releases_url = reverse("browse-origin-releases", query_params=query_params) else: assert snapshot_id is not None branches, releases, aliases = get_snapshot_content(snapshot_id) url_args = {"snapshot_id": snapshot_id} directory_url = reverse("browse-snapshot-directory", 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_sizes_cache_id = f"swh_snapshot_{snapshot_id}_sizes" snapshot_sizes = cache.get(snapshot_sizes_cache_id) if snapshot_sizes is None: snapshot_sizes = archive.lookup_snapshot_sizes(snapshot_id) cache.set(snapshot_sizes_cache_id, snapshot_sizes) is_empty = (snapshot_sizes["release"] + snapshot_sizes["revision"]) == 0 swh_snp_id = str( CoreSWHID(object_type=ObjectType.SNAPSHOT, object_id=hash_to_bytes(snapshot_id)) ) if visit_info: timestamp = format_utc_iso_date(visit_info["date"]) if origin_info: browse_view_name = f"browse-origin-{browse_context}" else: browse_view_name = f"browse-snapshot-{browse_context}" release_id = None root_directory = None snapshot_total_size = snapshot_sizes["release"] + snapshot_sizes["revision"] if path is not None: query_params["path"] = path if snapshot_total_size and revision_id is not None: # browse specific revision for a snapshot requested revision = archive.lookup_revision(revision_id) root_directory = revision["directory"] branches.append( SnapshotBranchInfo( name=revision_id, alias=False, revision=revision_id, directory=root_directory, date=revision["date"], message=revision["message"], url=None, ) ) query_params["revision"] = revision_id elif snapshot_total_size and release_name: # browse specific release for a snapshot requested release = _get_release(releases, release_name, snapshot_id) if release is None: _branch_not_found( "release", release_name, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id, ) else: if release["target_type"] == "revision": revision = archive.lookup_revision(release["target"]) root_directory = revision["directory"] revision_id = release["target"] elif release["target_type"] == "directory": root_directory = release["target"] release_id = release["id"] query_params["release"] = release_name elif snapshot_total_size: head = aliases.get("HEAD") if branch_name: # browse specific branch for a snapshot requested query_params["branch"] = branch_name branch = _get_branch(branches, branch_name, snapshot_id) if branch is None: _branch_not_found( "branch", branch_name, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id, ) else: branch_name = branch["name"] revision_id = branch["revision"] root_directory = branch["directory"] elif head is not None: # otherwise, browse branch targeted by the HEAD alias if it exists if head["target_type"] == "revision": # HEAD alias targets a revision head_rev = archive.lookup_revision(head["target"]) branch_name = "HEAD" revision_id = head_rev["id"] root_directory = head_rev["directory"] else: # HEAD alias targets a release release_name = archive.lookup_release(head["target"])["name"] head_rel = _get_release(releases, release_name, snapshot_id) if head_rel["target_type"] == "revision": revision = archive.lookup_revision(head_rel["target"]) root_directory = revision["directory"] revision_id = head_rel["target"] elif head_rel["target_type"] == "directory": root_directory = head_rel["target"] release_id = head_rel["id"] elif branches: # fallback to browse first branch otherwise branch = branches[0] branch_name = branch["name"] revision_id = branch["revision"] root_directory = branch["directory"] elif releases: # fallback to browse last release otherwise release = releases[-1] if release["target_type"] == "revision": revision = archive.lookup_revision(release["target"]) root_directory = revision["directory"] revision_id = release["target"] elif release["target_type"] == "directory": root_directory = release["target"] release_id = release["id"] release_name = release["name"] for b in branches: branch_query_params = dict(query_params) branch_query_params.pop("release", None) if b["name"] != b["revision"]: branch_query_params.pop("revision", None) branch_query_params["branch"] = b["name"] b["url"] = reverse( browse_view_name, url_args=url_args, query_params=branch_query_params ) for r in releases: release_query_params = dict(query_params) release_query_params.pop("branch", None) release_query_params.pop("revision", None) release_query_params["release"] = r["name"] r["url"] = reverse( browse_view_name, url_args=url_args, query_params=release_query_params, ) revision_info = None if revision_id: try: revision_info = archive.lookup_revision(revision_id) except NotFoundExc: pass else: revision_info["date"] = format_utc_iso_date(revision_info["date"]) revision_info["committer_date"] = format_utc_iso_date( revision_info["committer_date"] ) if revision_info["message"]: message_lines = revision_info["message"].split("\n") revision_info["message_header"] = message_lines[0] else: revision_info["message_header"] = "" snapshot_context = SnapshotContext( directory_url=directory_url, branch=branch_name, branch_alias=branch_name in aliases, branches=branches, branches_url=branches_url, is_empty=is_empty, origin_info=origin_info, origin_visits_url=origin_visits_url, release=release_name, release_alias=release_name in aliases, release_id=release_id, query_params=query_params, releases=releases, releases_url=releases_url, revision_id=revision_id, revision_info=revision_info, root_directory=root_directory, snapshot_id=snapshot_id, snapshot_sizes=snapshot_sizes, snapshot_swhid=swh_snp_id, url_args=url_args, visit_info=visit_info, ) if revision_info: revision_info["revision_url"] = gen_revision_url(revision_id, snapshot_context) return snapshot_context def _build_breadcrumbs(snapshot_context: SnapshotContext, path: str): origin_info = snapshot_context["origin_info"] url_args = snapshot_context["url_args"] query_params = dict(snapshot_context["query_params"]) root_directory = snapshot_context["root_directory"] path_info = gen_path_info(path) if origin_info: browse_view_name = "browse-origin-directory" else: browse_view_name = "browse-snapshot-directory" breadcrumbs = [] if root_directory: query_params.pop("path", None) breadcrumbs.append( { "name": root_directory[:7], "url": reverse( browse_view_name, url_args=url_args, query_params=query_params ), } ) for pi in path_info: query_params["path"] = pi["path"] breadcrumbs.append( { "name": pi["name"], "url": reverse( browse_view_name, url_args=url_args, query_params=query_params ), } ) return breadcrumbs def _check_origin_url(snapshot_id, origin_url): if snapshot_id is None and origin_url is None: raise BadInputExc("An origin URL must be provided as query parameter.") def browse_snapshot_directory( request, snapshot_id=None, origin_url=None, timestamp=None, path=None ): """ Django view implementation for browsing a directory in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) + visit_id = int(request.GET.get("visit_id", 0)) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, - visit_id=request.GET.get("visit_id"), + visit_id=visit_id or None, path=path, browse_context="directory", branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), ) root_directory = snapshot_context["root_directory"] sha1_git = root_directory error_info = { "status_code": 200, "description": None, } if root_directory and path: try: dir_info = archive.lookup_directory_with_path(root_directory, path) sha1_git = dir_info["target"] except NotFoundExc as e: sha1_git = None error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" dirs = [] files = [] if sha1_git: dirs, files = get_directory_entries(sha1_git) origin_info = snapshot_context["origin_info"] visit_info = snapshot_context["visit_info"] url_args = snapshot_context["url_args"] query_params = dict(snapshot_context["query_params"]) revision_id = snapshot_context["revision_id"] snapshot_id = snapshot_context["snapshot_id"] if origin_info: browse_view_name = "browse-origin-directory" else: browse_view_name = "browse-snapshot-directory" breadcrumbs = _build_breadcrumbs(snapshot_context, path) 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: query_params["path"] = path + d["name"] d["url"] = reverse( browse_view_name, url_args=url_args, query_params=query_params ) sum_file_sizes = 0 readmes = {} if origin_info: browse_view_name = "browse-origin-content" else: browse_view_name = "browse-snapshot-content" for f in files: query_params["path"] = path + f["name"] f["url"] = reverse( browse_view_name, url_args=url_args, query_params=query_params ) if f["length"] is not None: sum_file_sizes += 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) if origin_info: browse_view_name = "browse-origin-log" else: browse_view_name = "browse-snapshot-log" history_url = None if snapshot_id != _empty_snapshot_id: query_params.pop("path", None) history_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) nb_files = None nb_dirs = None dir_path = None if root_directory: nb_files = len(files) nb_dirs = len(dirs) dir_path = "/" + path swh_objects = [] vault_cooking = {} revision_found = False if revision_id is not None: try: archive.lookup_revision(revision_id) except NotFoundExc: pass else: revision_found = True if sha1_git is not None: swh_objects.append( SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=sha1_git) ) vault_cooking.update( {"directory_context": True, "directory_swhid": f"swh:1:dir:{sha1_git}",} ) if revision_id is not None and revision_found: swh_objects.append( SWHObjectInfo(object_type=ObjectType.REVISION, object_id=revision_id) ) vault_cooking.update( {"revision_context": True, "revision_swhid": f"swh:1:rev:{revision_id}",} ) swh_objects.append( SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id) ) visit_date = None visit_type = None if visit_info: visit_date = format_utc_iso_date(visit_info["date"]) visit_type = visit_info["type"] release_id = snapshot_context["release_id"] if release_id: swh_objects.append( SWHObjectInfo(object_type=ObjectType.RELEASE, object_id=release_id) ) dir_metadata = DirectoryMetadata( object_type=ObjectType.DIRECTORY, object_id=sha1_git, directory=sha1_git, nb_files=nb_files, nb_dirs=nb_dirs, sum_file_sizes=sum_file_sizes, root_directory=root_directory, path=dir_path, revision=revision_id, revision_found=revision_found, release=release_id, snapshot=snapshot_id, origin_url=origin_url, visit_date=visit_date, visit_type=visit_type, ) swhids_info = get_swhids_info(swh_objects, snapshot_context, dir_metadata) 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, ) top_right_link = None if not snapshot_context["is_empty"] and revision_found: top_right_link = { "url": history_url, "icon": swh_object_icons["revisions history"], "text": "History", } 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_directory else [], "top_right_link": top_right_link, "readme_name": readme_name, "readme_url": readme_url, "readme_html": readme_html, "snapshot_context": snapshot_context, "vault_cooking": vault_cooking, "show_actions": True, "swhids_info": swhids_info, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], }, status=error_info["status_code"], ) PER_PAGE = 100 def browse_snapshot_log(request, snapshot_id=None, origin_url=None, timestamp=None): """ Django view implementation for browsing a revision history in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) + visit_id = int(request.GET.get("visit_id", 0)) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, - visit_id=request.GET.get("visit_id"), + visit_id=visit_id or None, browse_context="log", branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), ) revision_id = snapshot_context["revision_id"] if revision_id is None: raise NotFoundExc("No revisions history found in the current snapshot context.") 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 = archive.get_revisions_walker( revs_ordering, revision_id, max_revs=offset + per_page + 1, state=revs_walker_state, ) rev_log += [rev["id"] for rev in revs_walker] revs_walker_state = revs_walker.export_state() revs = rev_log[offset : offset + per_page] revision_log = archive.lookup_revision_multiple(revs) request.session[session_key] = { "rev_log": rev_log, "revs_walker_state": revs_walker_state, } 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 or None if origin_info: browse_view_name = "browse-origin-log" else: browse_view_name = "browse-snapshot-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_rev_link = gen_revision_link(revision_id) browse_log_link = gen_revision_log_link(revision_id) browse_snp_link = gen_snapshot_link(snapshot_id) revision_metadata = { "context-independent revision": browse_rev_link, "context-independent revision history": browse_log_link, "context-independent snapshot": browse_snp_link, "snapshot": snapshot_id, } if origin_info: revision_metadata["origin url"] = origin_info["url"] revision_metadata["origin visit date"] = format_utc_iso_date(visit_info["date"]) revision_metadata["origin visit type"] = visit_info["type"] swh_objects = [ SWHObjectInfo(object_type=ObjectType.REVISION, object_id=revision_id), SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id), ] release_id = snapshot_context["release_id"] if release_id: swh_objects.append( SWHObjectInfo(object_type=ObjectType.RELEASE, object_id=release_id) ) browse_rel_link = gen_release_link(release_id) revision_metadata["release"] = release_id revision_metadata["context-independent release"] = browse_rel_link swhids_info = get_swhids_info(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": True, "swhids_info": swhids_info, }, ) def browse_snapshot_branches( request, snapshot_id=None, origin_url=None, timestamp=None, branch_name_include=None ): """ Django view implementation for browsing a list of branches in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) + visit_id = int(request.GET.get("visit_id", 0)) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, - visit_id=request.GET.get("visit_id"), + visit_id=visit_id or None, ) 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 "" origin_info = snapshot_context["origin_info"] url_args = snapshot_context["url_args"] query_params = snapshot_context["query_params"] if origin_info: browse_view_name = "browse-origin-directory" else: browse_view_name = "browse-snapshot-directory" snapshot = archive.lookup_snapshot( snapshot_context["snapshot_id"], branches_from, PER_PAGE + 1, target_types=["revision", "alias"], branch_name_include_substring=branch_name_include, ) displayed_branches = [] if snapshot: displayed_branches, _, _ = process_snapshot_branches(snapshot) for branch in displayed_branches: rev_query_params = {} if origin_info: rev_query_params["origin_url"] = origin_info["url"] revision_url = reverse( "browse-revision", url_args={"sha1_git": branch["revision"]}, query_params=query_params, ) 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 if origin_info: browse_view_name = "browse-origin-branches" else: browse_view_name = "browse-snapshot-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 snapshot and snapshot["next_branch"] is not None: 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, "search_string": branch_name_include or "", }, ) def browse_snapshot_releases( request, snapshot_id=None, origin_url=None, timestamp=None, release_name_include=None, ): """ Django view implementation for browsing a list of releases in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) + visit_id = int(request.GET.get("visit_id", 0)) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, - visit_id=request.GET.get("visit_id"), + visit_id=visit_id or None, ) 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 "" origin_info = snapshot_context["origin_info"] url_args = snapshot_context["url_args"] query_params = snapshot_context["query_params"] snapshot = archive.lookup_snapshot( snapshot_context["snapshot_id"], rel_from, PER_PAGE + 1, target_types=["release", "alias"], branch_name_include_substring=release_name_include, ) displayed_releases = [] if snapshot: _, displayed_releases, _ = process_snapshot_branches(snapshot) for release in displayed_releases: query_params_tgt = {"snapshot": snapshot_id, "release": release["name"]} if origin_info: query_params_tgt["origin_url"] = origin_info["url"] release_url = reverse( "browse-release", url_args={"sha1_git": release["id"]}, query_params=query_params_tgt, ) target_url = "" tooltip = ( f"The release {release['name']} targets " f"{release['target_type']} {release['target']}" ) 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={"query_string": 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, ) tooltip = ( f"The release {release['name']} " f"is an alias for release {release['target']}" ) release["release_url"] = release_url release["target_url"] = target_url release["tooltip"] = tooltip if origin_info: browse_view_name = "browse-origin-releases" else: browse_view_name = "browse-snapshot-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 snapshot and snapshot["next_branch"] is not None: 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": False, "search_string": release_name_include or "", }, ) diff --git a/swh/web/browse/views/content.py b/swh/web/browse/views/content.py index 97897595..84d85225 100644 --- a/swh/web/browse/views/content.py +++ b/swh/web/browse/views/content.py @@ -1,444 +1,447 @@ -# Copyright (C) 2017-2021 The Software Heritage developers +# Copyright (C) 2017-2022 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 from distutils.util import strtobool import sentry_sdk from django.http import HttpResponse, JsonResponse from django.shortcuts import redirect, render from swh.model.hashutil import hash_to_hex from swh.model.swhids import ObjectType from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import ( content_display_max_size, gen_link, prepare_content_for_display, request_content, ) from swh.web.common import archive, highlightjs, query from swh.web.common.exc import BadInputExc, NotFoundExc, http_status_code_message from swh.web.common.identifiers import get_swhids_info from swh.web.common.typing import ContentMetadata, SWHObjectInfo from swh.web.common.utils import gen_path_info, reverse, swh_object_icons @browse_route( r"content/(?P[0-9a-z_:]*[0-9a-f]+.)/raw/", view_name="browse-content-raw", checksum_args=["query_string"], ) def content_raw(request, query_string): """Django view that produces a raw display of a content identified by its hash value. The url that points to it is :http:get:`/browse/content/[(algo_hash):](hash)/raw/` """ re_encode = bool(strtobool(request.GET.get("re_encode", "false"))) algo, checksum = query.parse_hash(query_string) checksum = hash_to_hex(checksum) content_data = request_content(query_string, max_size=None, re_encode=re_encode) 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.*)/", 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 = "plaintext" 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 = "plaintext" elif not force and diff_size > _auto_diff_size_limit: diff_str = "Large diffs are not automatically computed" language = "plaintext" 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 exc: sentry_sdk.capture_exception(exc) diff_str = str(exc) diff_data["diff_str"] = diff_str diff_data["language"] = language return JsonResponse(diff_data) def _get_content_from_request(request): path = request.GET.get("path") if path is None: raise BadInputExc("The path query parameter must be provided.") snapshot = request.GET.get("snapshot") or request.GET.get("snapshot_id") origin_url = request.GET.get("origin_url") if snapshot is None and origin_url is None: raise BadInputExc( "The origin_url or snapshot query parameters must be provided." ) + + visit_id = int(request.GET.get("visit_id", 0)) snapshot_context = get_snapshot_context( snapshot_id=snapshot, origin_url=origin_url, path=path, timestamp=request.GET.get("timestamp"), - visit_id=request.GET.get("visit_id"), + visit_id=visit_id or None, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), browse_context="content", ) root_directory = snapshot_context["root_directory"] return archive.lookup_directory_with_path(root_directory, path) @browse_route( r"content/(?P[0-9a-z_:]*[0-9a-f]+.)/", r"content/", view_name="browse-content", checksum_args=["query_string"], ) def content_display(request, query_string=None): """Django view that produces an HTML display of a content identified by its hash value. The URLs that points to it are :http:get:`/browse/content/[(algo_hash):](hash)/` :http:get:`/browse/content/` """ if query_string is None: # this case happens when redirected from origin/content or snapshot/content content = _get_content_from_request(request) return redirect( reverse( "browse-content", url_args={"query_string": f"sha1_git:{content['target']}"}, query_params=request.GET, ), ) algo, checksum = query.parse_hash(query_string) checksum = hash_to_hex(checksum) origin_url = request.GET.get("origin_url") selected_language = request.GET.get("language") if not origin_url: origin_url = request.GET.get("origin") snapshot_id = request.GET.get("snapshot") or request.GET.get("snapshot_id") path = request.GET.get("path") content_data = {} error_info = {"status_code": 200, "description": None} try: content_data = request_content(query_string) except NotFoundExc as e: error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" snapshot_context = None if origin_url is not None or snapshot_id is not None: try: + visit_id = int(request.GET.get("visit_id", 0)) snapshot_context = get_snapshot_context( origin_url=origin_url, snapshot_id=snapshot_id, timestamp=request.GET.get("timestamp"), - visit_id=request.GET.get("visit_id"), + visit_id=visit_id or None, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), path=path, browse_context="content", ) except NotFoundExc as e: if str(e).startswith("Origin"): raw_cnt_url = reverse( "browse-content", url_args={"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) else: raise e content = None language = None mimetype = None if content_data.get("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"] # Override language with user-selected language if selected_language is not None: language = selected_language available_languages = None if mimetype and "text/" in mimetype: available_languages = highlightjs.get_supported_languages() filename = None path_info = None directory_id = None root_dir = None if snapshot_context: root_dir = snapshot_context.get("root_directory") query_params = snapshot_context["query_params"] if snapshot_context else {} breadcrumbs = [] if path: split_path = path.split("/") root_dir = root_dir or split_path[0] filename = split_path[-1] if root_dir != path: path = path.replace(root_dir + "/", "") path = path[: -len(filename)] path_info = gen_path_info(path) query_params.pop("path", None) dir_url = reverse( "browse-directory", url_args={"sha1_git": root_dir}, query_params=query_params, ) breadcrumbs.append({"name": root_dir[:7], "url": dir_url}) for pi in path_info: query_params["path"] = pi["path"] dir_url = reverse( "browse-directory", url_args={"sha1_git": root_dir}, query_params=query_params, ) breadcrumbs.append({"name": pi["name"], "url": dir_url}) breadcrumbs.append({"name": filename, "url": None}) if path and root_dir != path: dir_info = archive.lookup_directory_with_path(root_dir, path) directory_id = dir_info["target"] elif root_dir != path: directory_id = root_dir else: root_dir = None query_params = {"filename": filename} content_checksums = content_data.get("checksums", {}) content_url = reverse("browse-content", url_args={"query_string": query_string},) content_raw_url = reverse( "browse-content-raw", url_args={"query_string": query_string}, query_params=query_params, ) content_metadata = ContentMetadata( object_type=ObjectType.CONTENT, object_id=content_checksums.get("sha1_git"), sha1=content_checksums.get("sha1"), sha1_git=content_checksums.get("sha1_git"), sha256=content_checksums.get("sha256"), blake2s256=content_checksums.get("blake2s256"), content_url=content_url, mimetype=content_data.get("mimetype"), encoding=content_data.get("encoding"), size=content_data.get("length", 0), language=content_data.get("language"), root_directory=root_dir, path=f"/{path}" if path else None, filename=filename or "", directory=directory_id, revision=None, release=None, snapshot=None, origin_url=origin_url, ) swh_objects = [] if content_checksums: swh_objects.append( SWHObjectInfo( object_type=ObjectType.CONTENT, object_id=content_checksums.get("sha1_git"), ) ) if directory_id: swh_objects.append( SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=directory_id) ) if snapshot_context: if snapshot_context["revision_id"]: swh_objects.append( SWHObjectInfo( object_type=ObjectType.REVISION, object_id=snapshot_context["revision_id"], ) ) swh_objects.append( SWHObjectInfo( object_type=ObjectType.SNAPSHOT, object_id=snapshot_context["snapshot_id"], ) ) if snapshot_context["release_id"]: swh_objects.append( SWHObjectInfo( object_type=ObjectType.RELEASE, object_id=snapshot_context["release_id"], ) ) swhids_info = get_swhids_info( swh_objects, snapshot_context, extra_context=content_metadata, ) heading = "Content - %s" % content_checksums.get("sha1_git") if breadcrumbs: content_path = "/".join([bc["name"] for bc in breadcrumbs]) heading += " - %s" % content_path return render( request, "browse/content.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"] if swhids_info else "", "swh_object_name": "Content", "swh_object_metadata": content_metadata, "content": content, "content_size": content_data.get("length"), "max_content_size": content_display_max_size, "filename": filename, "encoding": content_data.get("encoding"), "mimetype": mimetype, "language": language, "available_languages": available_languages, "breadcrumbs": breadcrumbs, "top_right_link": { "url": content_raw_url, "icon": swh_object_icons["content"], "text": "Raw File", }, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": True, "swhids_info": swhids_info, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], }, status=error_info["status_code"], ) diff --git a/swh/web/browse/views/release.py b/swh/web/browse/views/release.py index 99f8a765..e9faaa94 100644 --- a/swh/web/browse/views/release.py +++ b/swh/web/browse/views/release.py @@ -1,245 +1,245 @@ -# Copyright (C) 2017-2021 The Software Heritage developers +# Copyright (C) 2017-2022 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 sentry_sdk from django.shortcuts import render from swh.model.swhids import ObjectType from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import ( gen_content_link, gen_directory_link, gen_link, gen_person_mail_link, gen_release_link, gen_revision_link, ) from swh.web.common import archive from swh.web.common.exc import NotFoundExc from swh.web.common.identifiers import get_swhids_info from swh.web.common.typing import ReleaseMetadata, SWHObjectInfo from swh.web.common.utils import format_utc_iso_date, reverse @browse_route( r"release/(?P[0-9a-f]+)/", view_name="browse-release", checksum_args=["sha1_git"], ) def release_browse(request, sha1_git): """ Django view that produces an HTML display of a release identified by its id. The url that points to it is :http:get:`/browse/release/(sha1_git)/`. """ release = archive.lookup_release(sha1_git) snapshot_context = {} origin_info = None snapshot_id = request.GET.get("snapshot_id") if not snapshot_id: snapshot_id = request.GET.get("snapshot") origin_url = request.GET.get("origin_url") if not origin_url: origin_url = request.GET.get("origin") timestamp = request.GET.get("timestamp") - visit_id = request.GET.get("visit_id") + visit_id = int(request.GET.get("visit_id", 0)) if origin_url: try: snapshot_context = get_snapshot_context( snapshot_id, origin_url, timestamp, - visit_id, + visit_id or None, release_name=release["name"], ) except NotFoundExc as e: raw_rel_url = reverse("browse-release", url_args={"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)) ) if str(e).startswith("Origin"): raise NotFoundExc(error_message) else: raise e origin_info = snapshot_context["origin_info"] elif snapshot_id: snapshot_context = get_snapshot_context( snapshot_id, release_name=release["name"] ) snapshot_id = snapshot_context.get("snapshot_id", None) release_metadata = ReleaseMetadata( object_type=ObjectType.RELEASE, object_id=sha1_git, release=sha1_git, author=release["author"]["fullname"] if release["author"] else "None", author_url=gen_person_mail_link(release["author"]) if release["author"] else "None", date=format_utc_iso_date(release["date"]), name=release["name"], synthetic=release["synthetic"], target=release["target"], target_type=release["target_type"], snapshot=snapshot_id, origin_url=origin_url, ) release_note_lines = [] if release["message"]: release_note_lines = release["message"].split("\n") swh_objects = [SWHObjectInfo(object_type=ObjectType.RELEASE, object_id=sha1_git)] vault_cooking = None rev_directory = None target_link = None if release["target_type"] == ObjectType.REVISION.name.lower(): target_link = gen_revision_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) try: revision = archive.lookup_revision(release["target"]) rev_directory = revision["directory"] vault_cooking = { "directory_context": True, "directory_swhid": f"swh:1:dir:{rev_directory}", "revision_context": True, "revision_swhid": f"swh:1:rev:{release['target']}", } swh_objects.append( SWHObjectInfo( object_type=ObjectType.REVISION, object_id=release["target"] ) ) swh_objects.append( SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=rev_directory) ) except Exception as exc: sentry_sdk.capture_exception(exc) elif release["target_type"] == ObjectType.DIRECTORY.name.lower(): target_link = gen_directory_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) try: # check directory exists archive.lookup_directory(release["target"]) vault_cooking = { "directory_context": True, "directory_swhid": f"swh:1:dir:{release['target']}", "revision_context": False, "revision_swhid": None, } swh_objects.append( SWHObjectInfo( object_type=ObjectType.DIRECTORY, object_id=release["target"] ) ) except Exception as exc: sentry_sdk.capture_exception(exc) elif release["target_type"] == ObjectType.CONTENT.name.lower(): target_link = gen_content_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) swh_objects.append( SWHObjectInfo(object_type=ObjectType.CONTENT, object_id=release["target"]) ) elif release["target_type"] == ObjectType.RELEASE.name.lower(): target_link = gen_release_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) rev_directory_url = None if rev_directory is not None: if origin_info: rev_directory_url = reverse( "browse-origin-directory", query_params={ "origin_url": origin_info["url"], "release": release["name"], "snapshot": snapshot_id, }, ) elif snapshot_id: rev_directory_url = reverse( "browse-snapshot-directory", url_args={"snapshot_id": snapshot_id}, query_params={"release": release["name"]}, ) else: rev_directory_url = reverse( "browse-directory", url_args={"sha1_git": rev_directory} ) directory_link = None if rev_directory_url is not None: directory_link = gen_link(rev_directory_url, rev_directory) release["directory_link"] = directory_link release["target_link"] = target_link if snapshot_context: snapshot_id = snapshot_context["snapshot_id"] if snapshot_id: swh_objects.append( SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id) ) swhids_info = get_swhids_info(swh_objects, snapshot_context) note_header = "None" if len(release_note_lines) > 0: note_header = release_note_lines[0] release["note_header"] = note_header release["note_body"] = "\n".join(release_note_lines[1:]) 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, "browse/release.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"], "swh_object_name": "Release", "swh_object_metadata": release_metadata, "release": release, "snapshot_context": snapshot_context, "show_actions": True, "breadcrumbs": None, "vault_cooking": vault_cooking, "top_right_link": None, "swhids_info": swhids_info, }, ) diff --git a/swh/web/browse/views/revision.py b/swh/web/browse/views/revision.py index 66ace89f..bbaab1ed 100644 --- a/swh/web/browse/views/revision.py +++ b/swh/web/browse/views/revision.py @@ -1,586 +1,587 @@ -# Copyright (C) 2017-2021 The Software Heritage developers +# Copyright (C) 2017-2022 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 JsonResponse from django.shortcuts import render from django.utils.safestring import mark_safe from swh.model.hashutil import hash_to_bytes from swh.model.swhids import CoreSWHID, ObjectType from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import ( content_display_max_size, format_log_entries, gen_link, gen_person_mail_link, gen_revision_url, get_directory_entries, get_readme_to_display, get_revision_log_url, prepare_content_for_display, request_content, ) from swh.web.common import archive from swh.web.common.exc import NotFoundExc, http_status_code_message from swh.web.common.identifiers import get_swhids_info from swh.web.common.typing import RevisionMetadata, SWHObjectInfo from swh.web.common.utils import ( format_utc_iso_date, gen_path_info, reverse, swh_object_icons, ) def _gen_content_url(revision, query_string, path, snapshot_context): if snapshot_context: query_params = snapshot_context["query_params"] query_params["path"] = path query_params["revision"] = revision["id"] content_url = reverse("browse-origin-content", query_params=query_params) else: content_path = "%s/%s" % (revision["directory"], path) content_url = reverse( "browse-content", url_args={"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", url_args=url_args, query_params=query_params ) hasher.update(diff_id.encode("utf-8")) diff_id = hasher.hexdigest() change["id"] = diff_id diff_link = "#diff_" + 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, 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, diff_link, change["to_path"]) ) elif change["type"] == "delete": parent = archive.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, 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, 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", checksum_args=["sha1_git"], ) def _revision_diff(request, sha1_git): """ Browse internal endpoint to compute revision diff """ revision = archive.lookup_revision(sha1_git) snapshot_context = 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( origin_url=origin_url, timestamp=timestamp, visit_id=visit_id ) changes = archive.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, } return JsonResponse(diff_data) NB_LOG_ENTRIES = 100 @browse_route( r"revision/(?P[0-9a-f]+)/log/", view_name="browse-revision-log", checksum_args=["sha1_git"], ) def revision_log_browse(request, sha1_git): """ Django view that produces an HTML display of the history log for a revision identified by its id. The url that points to it is :http:get:`/browse/revision/(sha1_git)/log/` """ origin_url = request.GET.get("origin_url") snapshot_id = request.GET.get("snapshot") snapshot_context = None if origin_url or snapshot_id: + visit_id = int(request.GET.get("visit_id", 0)) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=request.GET.get("timestamp"), - visit_id=request.GET.get("visit_id"), + visit_id=visit_id or None, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=sha1_git, ) per_page = int(request.GET.get("per_page", NB_LOG_ENTRIES)) offset = int(request.GET.get("offset", 0)) revs_ordering = request.GET.get("revs_ordering", "committer_date") session_key = "rev_%s_log_ordering_%s" % (sha1_git, 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 = archive.get_revisions_walker( revs_ordering, sha1_git, max_revs=offset + per_page + 1, state=revs_walker_state, ) rev_log += [rev["id"] for rev in revs_walker] revs_walker_state = revs_walker.export_state() revs = rev_log[offset : offset + per_page] revision_log = archive.lookup_revision_multiple(revs) request.session[session_key] = { "rev_log": rev_log, "revs_walker_state": revs_walker_state, } revs_ordering = request.GET.get("revs_ordering", "") prev_log_url = None if len(rev_log) > offset + per_page: prev_log_url = reverse( "browse-revision-log", url_args={"sha1_git": sha1_git}, query_params={ "per_page": per_page, "offset": offset + per_page, "revs_ordering": revs_ordering or None, }, ) next_log_url = None if offset != 0: next_log_url = reverse( "browse-revision-log", url_args={"sha1_git": sha1_git}, query_params={ "per_page": per_page, "offset": offset - per_page, "revs_ordering": revs_ordering or None, }, ) revision_log_data = format_log_entries(revision_log, per_page) swh_rev_id = str( CoreSWHID(object_type=ObjectType.REVISION, object_id=hash_to_bytes(sha1_git)) ) return render( request, "browse/revision-log.html", { "heading": "Revision history", "swh_object_id": swh_rev_id, "swh_object_name": "Revisions history", "swh_object_metadata": None, "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": True, "swhids_info": None, }, ) @browse_route( r"revision/(?P[0-9a-f]+)/", view_name="browse-revision", checksum_args=["sha1_git"], ) def revision_browse(request, sha1_git): """ Django view that produces an HTML display of a revision identified by its id. The url that points to it is :http:get:`/browse/revision/(sha1_git)/`. """ revision = archive.lookup_revision(sha1_git) origin_info = None snapshot_context = None origin_url = request.GET.get("origin_url") if not origin_url: origin_url = request.GET.get("origin") timestamp = request.GET.get("timestamp") - visit_id = request.GET.get("visit_id") + visit_id = int(request.GET.get("visit_id", 0)) snapshot_id = request.GET.get("snapshot_id") if not snapshot_id: snapshot_id = request.GET.get("snapshot") path = request.GET.get("path") dir_id = None dirs, files = [], [] content_data = {} if origin_url: try: snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, - visit_id=visit_id, + visit_id=visit_id or None, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=sha1_git, path=path, ) except NotFoundExc as e: raw_rev_url = reverse("browse-revision", url_args={"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)) ) if str(e).startswith("Origin"): raise NotFoundExc(error_message) else: raise e origin_info = snapshot_context["origin_info"] snapshot_id = snapshot_context["snapshot_id"] elif snapshot_id: snapshot_context = get_snapshot_context(snapshot_id) error_info = {"status_code": 200, "description": None} if path: try: file_info = archive.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) except NotFoundExc as e: error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" else: dir_id = revision["directory"] if dir_id: path = "" if path is None else (path + "/") dirs, files = get_directory_entries(dir_id) revision_metadata = RevisionMetadata( object_type=ObjectType.REVISION, object_id=sha1_git, revision=sha1_git, author=revision["author"]["fullname"] if revision["author"] else "None", author_url=gen_person_mail_link(revision["author"]) if revision["author"] else "None", committer=revision["committer"]["fullname"] if revision["committer"] else "None", committer_url=gen_person_mail_link(revision["committer"]) if revision["committer"] else "None", committer_date=format_utc_iso_date(revision["committer_date"]), date=format_utc_iso_date(revision["date"]), directory=revision["directory"], merge=revision["merge"], metadata=json.dumps( revision["metadata"], sort_keys=True, indent=4, separators=(",", ": ") ), parents=revision["parents"], synthetic=revision["synthetic"], type=revision["type"], snapshot=snapshot_id, origin_url=origin_url, ) message_lines = ["None"] if revision["message"]: message_lines = revision["message"].split("\n") parents = [] for p in revision["parents"]: parent_url = gen_revision_url(p, snapshot_context) parents.append({"id": p, "url": parent_url}) path_info = gen_path_info(path) query_params = snapshot_context["query_params"] if snapshot_context else {} breadcrumbs = [] breadcrumbs.append( { "name": revision["directory"][:7], "url": reverse( "browse-revision", url_args={"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", url_args={"sha1_git": sha1_git}, query_params=query_params, ), } ) vault_cooking = { "directory_context": False, "directory_swhid": None, "revision_context": True, "revision_swhid": f"swh:1:rev:{sha1_git}", } swh_objects = [SWHObjectInfo(object_type=ObjectType.REVISION, object_id=sha1_git)] content = None content_size = None filename = None mimetype = None language = None readme_name = None readme_url = None readme_html = None readmes = {} extra_context = dict(revision_metadata) extra_context["path"] = f"/{path}" if path else None 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"] mimetype = content_display_data["mimetype"] if path: filename = path_info[-1]["name"] query_params["filename"] = filename filepath = "/".join(pi["name"] for pi in path_info[:-1]) extra_context["path"] = f"/{filepath}/" if filepath else "/" extra_context["filename"] = filename top_right_link = { "url": reverse( "browse-content-raw", url_args={"query_string": query_string}, query_params={"filename": filename}, ), "icon": swh_object_icons["content"], "text": "Raw File", } swh_objects.append( SWHObjectInfo(object_type=ObjectType.CONTENT, object_id=file_info["target"]) ) else: for d in dirs: if d["type"] == "rev": d["url"] = reverse( "browse-revision", url_args={"sha1_git": d["target"]} ) else: query_params["path"] = path + d["name"] d["url"] = reverse( "browse-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ) for f in files: query_params["path"] = path + f["name"] f["url"] = reverse( "browse-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ) 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 = { "url": get_revision_log_url(sha1_git, snapshot_context), "icon": swh_object_icons["revisions history"], "text": "History", } vault_cooking["directory_context"] = True vault_cooking["directory_swhid"] = f"swh:1:dir:{dir_id}" swh_objects.append( SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=dir_id) ) query_params.pop("path", None) diff_revision_url = reverse( "diff-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ) if snapshot_id: swh_objects.append( SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id) ) swhids_info = get_swhids_info(swh_objects, snapshot_context, extra_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, "browse/revision.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"], "swh_object_name": "Revision", "swh_object_metadata": revision_metadata, "message_header": message_lines[0], "message_body": "\n".join(message_lines[1:]), "parents": parents, "snapshot_context": snapshot_context, "dirs": dirs, "files": files, "content": content, "content_size": content_size, "max_content_size": content_display_max_size, "filename": filename, "encoding": content_data.get("encoding"), "mimetype": mimetype, "language": language, "readme_name": readme_name, "readme_url": readme_url, "readme_html": readme_html, "breadcrumbs": breadcrumbs, "top_right_link": top_right_link, "vault_cooking": vault_cooking, "diff_revision_url": diff_revision_url, "show_actions": True, "swhids_info": swhids_info, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], }, status=error_info["status_code"], ) diff --git a/swh/web/common/archive.py b/swh/web/common/archive.py index cc095dcc..9a9869ec 100644 --- a/swh/web/common/archive.py +++ b/swh/web/common/archive.py @@ -1,1450 +1,1452 @@ # Copyright (C) 2015-2022 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 collections import defaultdict import datetime import itertools import os import re from typing import Any, Dict, Iterable, Iterator, List, Optional, Set, Tuple, Union from urllib.parse import urlparse from swh.model import hashutil from swh.model.model import OriginVisit, Revision from swh.model.swhids import CoreSWHID, ObjectType from swh.storage.algos import diff, revisions_walker from swh.storage.algos.origin import origin_get_latest_visit_status from swh.storage.algos.snapshot import snapshot_get_latest, snapshot_resolve_alias from swh.vault.exc import NotFoundExc as VaultNotFoundExc from swh.web import config from swh.web.common import converters, query from swh.web.common.exc import NotFoundExc from swh.web.common.typing import ( OriginInfo, OriginMetadataInfo, OriginVisitInfo, PagedResult, ) search = config.search() storage = config.storage() vault = config.vault() idx_storage = config.indexer_storage() counters = config.counters() MAX_LIMIT = 50 # Top limit the users can ask for def _first_element(lst): """Returns the first element in the provided list or None if it is empty or None""" return next(iter(lst or []), None) def lookup_multiple_hashes(hashes): """Lookup the passed hashes in a single DB connection, using batch processing. Args: An array of {filename: X, sha1: Y}, string X, hex sha1 string Y. Returns: The same array with elements updated with elem['found'] = true if the hash is present in storage, elem['found'] = false if not. """ hashlist = [hashutil.hash_to_bytes(elem["sha1"]) for elem in hashes] content_missing = storage.content_missing_per_sha1(hashlist) missing = [hashutil.hash_to_hex(x) for x in content_missing] for x in hashes: x.update({"found": True}) for h in hashes: if h["sha1"] in missing: h["found"] = False return hashes def lookup_expression(expression, last_sha1, per_page): """Lookup expression in raw content. Args: expression (str): An expression to lookup through raw indexed content last_sha1 (str): Last sha1 seen per_page (int): Number of results per page Yields: ctags whose content match the expression """ limit = min(per_page, MAX_LIMIT) ctags = idx_storage.content_ctags_search( expression, last_sha1=last_sha1, limit=limit ) for ctag in ctags: ctag = converters.from_swh(ctag, hashess={"id"}) ctag["sha1"] = ctag["id"] ctag.pop("id") yield ctag def lookup_hash(q: str) -> Dict[str, Any]: """Check if the storage contains a given content checksum and return it if found. Args: q: query string of the form Returns: Dict with key found containing the hash info if the hash is present, None if not. """ algo, hash_ = query.parse_hash(q) found = _first_element(storage.content_find({algo: hash_})) if found: content = converters.from_content(found.to_dict()) else: content = None return {"found": content, "algo": algo} def search_hash(q: str) -> Dict[str, bool]: """Search storage for a given content checksum. Args: q: query string of the form Returns: Dict with key found to True or False, according to whether the checksum is present or not """ algo, hash_ = query.parse_hash(q) found = _first_element(storage.content_find({algo: hash_})) return {"found": found is not None} def _lookup_content_sha1(q: str) -> Optional[bytes]: """Given a possible input, query for the content's sha1. Args: q: query string of the form Returns: binary sha1 if found or None """ algo, hash_ = query.parse_hash(q) if algo != "sha1": hashes = _first_element(storage.content_find({algo: hash_})) if not hashes: return None return hashes.sha1 return hash_ def lookup_content_ctags(q): """Return ctags information from a specified content. Args: q: query string of the form Yields: ctags information (dict) list if the content is found. """ sha1 = _lookup_content_sha1(q) if not sha1: return None ctags = list(idx_storage.content_ctags_get([sha1])) if not ctags: return None for ctag in ctags: yield converters.from_swh(ctag, hashess={"id"}) def lookup_content_filetype(q): """Return filetype information from a specified content. Args: q: query string of the form Yields: filetype information (dict) list if the content is found. """ sha1 = _lookup_content_sha1(q) if not sha1: return None filetype = _first_element(list(idx_storage.content_mimetype_get([sha1]))) if not filetype: return None return converters.from_filetype(filetype.to_dict()) def lookup_content_language(q): """Always returns None. This used to return language information from a specified content, but this is currently disabled. Args: q: query string of the form Yields: language information (dict) list if the content is found. """ return None def lookup_content_license(q): """Return license information from a specified content. Args: q: query string of the form Yields: license information (dict) list if the content is found. """ sha1 = _lookup_content_sha1(q) if not sha1: return None licenses = list(idx_storage.content_fossology_license_get([sha1])) if not licenses: return None license_dicts = [license.to_dict() for license in licenses] for license_dict in license_dicts: del license_dict["id"] lic = { "id": sha1, "facts": license_dicts, } return converters.from_swh(lic, hashess={"id"}) def lookup_origin(origin: OriginInfo) -> OriginInfo: """Return information about the origin matching dict origin. Args: origin: origin's dict with 'url' key Returns: origin information as dict. """ origin_urls = [origin["url"]] if origin["url"]: # handle case when user provided an origin url with a trailing # slash while the url in storage does not have it (e.g. GitHub) if origin["url"].endswith("/"): origin_urls.append(origin["url"][:-1]) # handle case when user provided an origin url without a trailing # slash while the url in storage have it (e.g. Debian source package) else: origin_urls.append(f"{origin['url']}/") try: # handle case where the "://" character sequence was mangled into ":/" parsed_url = urlparse(origin["url"]) if ( parsed_url.scheme and not parsed_url.netloc and origin["url"].startswith(f"{parsed_url.scheme}:/") and not origin["url"].startswith(f"{parsed_url.scheme}://") ): origin_urls.append( origin["url"].replace( f"{parsed_url.scheme}:/", f"{parsed_url.scheme}://" ) ) except Exception: pass origins = [o for o in storage.origin_get(origin_urls) if o is not None] if not origins: msg = "Origin with url %s not found!" % origin["url"] raise NotFoundExc(msg) return converters.from_origin(origins[0].to_dict()) def lookup_origins( page_token: Optional[str], limit: int = 100 ) -> PagedResult[OriginInfo]: """Get list of archived software origins in a paginated way. Origins are sorted by id before returning them Args: origin_from (int): The minimum id of the origins to return origin_count (int): The maximum number of origins to return Returns: Page of OriginInfo """ page = storage.origin_list(page_token=page_token, limit=limit) return PagedResult( [converters.from_origin(o.to_dict()) for o in page.results], next_page_token=page.next_page_token, ) def lookup_origin_snapshots(origin: OriginInfo) -> List[str]: """Return ids of the snapshots of an origin. Args: origin: origin's dict with 'url' key Returns: List of unique snapshot identifiers in hexadecimal format resulting from the visits of the origin. """ return [ snapshot.hex() for snapshot in storage.origin_snapshot_get_all(origin["url"]) ] def search_origin( url_pattern: str, use_ql: bool = False, limit: int = 50, with_visit: bool = False, visit_types: Optional[List[str]] = None, page_token: Optional[str] = None, ) -> Tuple[List[OriginInfo], Optional[str]]: """Search for origins whose urls contain a provided string pattern or match a provided regular expression. Args: url_pattern: the string pattern to search for in origin urls use_ql: whether to use swh search query language or not limit: the maximum number of found origins to return with_visit: Whether origins with no visit are to be filtered out visit_types: Only origins having any of the provided visit types (e.g. git, svn, pypi) will be returned page_token: opaque string used to get the next results of a search Returns: list of origin information as dict. """ if page_token: assert isinstance(page_token, str) if search: if use_ql: page_result = search.origin_search( query=url_pattern, page_token=page_token, with_visit=with_visit, visit_types=visit_types, limit=limit, ) else: page_result = search.origin_search( url_pattern=url_pattern, page_token=page_token, with_visit=with_visit, visit_types=visit_types, limit=limit, ) origins = [converters.from_origin(ori_dict) for ori_dict in page_result.results] else: # Fallback to swh-storage if swh-search is not configured search_words = [re.escape(word) for word in url_pattern.split()] if len(search_words) >= 7: url_pattern = ".*".join(search_words) else: pattern_parts = [] for permut in itertools.permutations(search_words): pattern_parts.append(".*".join(permut)) url_pattern = "|".join(pattern_parts) page_result = storage.origin_search( url_pattern, page_token=page_token, with_visit=with_visit, limit=limit, visit_types=visit_types, regexp=True, ) origins = [converters.from_origin(ori.to_dict()) for ori in page_result.results] return (origins, page_result.next_page_token) def search_origin_metadata( fulltext: str, limit: int = 50 ) -> Iterable[OriginMetadataInfo]: """Search for origins whose metadata match a provided string pattern. Args: fulltext: the string pattern to search for in origin metadata limit: the maximum number of found origins to return Returns: Iterable of origin metadata information for existing origins """ results = [] if ( search and config.get_config()["search_config"]["metadata_backend"] == "swh-search" ): page_result = search.origin_search(metadata_pattern=fulltext, limit=limit,) matches = idx_storage.origin_intrinsic_metadata_get( [r["url"] for r in page_result.results] ) else: matches = idx_storage.origin_intrinsic_metadata_search_fulltext( conjunction=[fulltext], limit=limit ) matches = [match.to_dict() for match in matches] origins = storage.origin_get([match["id"] for match in matches]) for origin, match in zip(origins, matches): if not origin: continue match["from_revision"] = hashutil.hash_to_hex(match["from_revision"]) del match["id"] results.append(OriginMetadataInfo(url=origin.url, metadata=match)) return results def lookup_origin_intrinsic_metadata(origin_url: str) -> Dict[str, Any]: """Return intrinsic metadata for origin whose origin matches given origin. Args: origin_url: origin url Raises: NotFoundExc when the origin is not found Returns: origin metadata. """ origins = [origin_url] origin_info = storage.origin_get(origins)[0] if not origin_info: raise NotFoundExc(f"Origin with url {origin_url} not found!") match = _first_element(idx_storage.origin_intrinsic_metadata_get(origins)) result = {} if match: result = match.metadata return result def _to_sha1_bin(sha1_hex): _, sha1_git_bin = query.parse_hash_with_algorithms_or_throws( sha1_hex, ["sha1"], "Only sha1_git is supported." # HACK: sha1_git really ) return sha1_git_bin def _check_directory_exists(sha1_git, sha1_git_bin): if len(list(storage.directory_missing([sha1_git_bin]))): raise NotFoundExc("Directory with sha1_git %s not found" % sha1_git) def lookup_directory(sha1_git): """Return information about the directory with id sha1_git. Args: sha1_git as string Returns: directory information as dict. """ empty_dir_sha1 = "4b825dc642cb6eb9a060e54bf8d69288fbee4904" if sha1_git == empty_dir_sha1: return [] sha1_git_bin = _to_sha1_bin(sha1_git) _check_directory_exists(sha1_git, sha1_git_bin) directory_entries = storage.directory_ls(sha1_git_bin) return map(converters.from_directory_entry, directory_entries) def lookup_directory_with_path(sha1_git: str, path: str) -> Dict[str, Any]: """Return directory information for entry with specified path w.r.t. root directory pointed by sha1_git Args: sha1_git: sha1_git corresponding to the directory to which we append paths to (hopefully) find the entry path: the relative path to the entry starting from the root directory pointed by sha1_git Returns: Directory entry information as dict. Raises: NotFoundExc if the directory entry is not found """ sha1_git_bin = _to_sha1_bin(sha1_git) _check_directory_exists(sha1_git, sha1_git_bin) paths = path.strip(os.path.sep).split(os.path.sep) queried_dir = storage.directory_entry_get_by_path( sha1_git_bin, [p.encode("utf-8") for p in paths] ) if not queried_dir: raise NotFoundExc( f"Directory entry with path {path} from root directory {sha1_git} not found" ) return converters.from_directory_entry(queried_dir) def lookup_release(release_sha1_git: str) -> Dict[str, Any]: """Return information about the release with sha1 release_sha1_git. Args: release_sha1_git: The release's sha1 as hexadecimal Returns: Release information as dict. Raises: ValueError if the identifier provided is not of sha1 nature. """ sha1_git_bin = _to_sha1_bin(release_sha1_git) release = _first_element(storage.release_get([sha1_git_bin])) if not release: raise NotFoundExc(f"Release with sha1_git {release_sha1_git} not found.") return converters.from_release(release) def lookup_release_multiple(sha1_git_list) -> Iterator[Optional[Dict[str, Any]]]: """Return information about the releases identified with their sha1_git identifiers. Args: sha1_git_list: A list of release sha1_git identifiers Returns: Iterator of Release metadata information as dict. Raises: ValueError if the identifier provided is not of sha1 nature. """ sha1_bin_list = [_to_sha1_bin(sha1_git) for sha1_git in sha1_git_list] releases = storage.release_get(sha1_bin_list) for r in releases: if r is not None: yield converters.from_release(r) else: yield None def lookup_revision(rev_sha1_git) -> Dict[str, Any]: """Return information about the revision with sha1 revision_sha1_git. Args: revision_sha1_git: The revision's sha1 as hexadecimal Returns: Revision information as dict. Raises: ValueError if the identifier provided is not of sha1 nature. NotFoundExc if there is no revision with the provided sha1_git. """ sha1_git_bin = _to_sha1_bin(rev_sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision with sha1_git {rev_sha1_git} not found.") return converters.from_revision(revision) def lookup_revision_multiple(sha1_git_list) -> Iterator[Optional[Dict[str, Any]]]: """Return information about the revisions identified with their sha1_git identifiers. Args: sha1_git_list: A list of revision sha1_git identifiers Yields: revision information as dict if the revision exists, None otherwise. Raises: ValueError if the identifier provided is not of sha1 nature. """ sha1_bin_list = [_to_sha1_bin(sha1_git) for sha1_git in sha1_git_list] revisions = storage.revision_get(sha1_bin_list) for revision in revisions: if revision is not None: yield converters.from_revision(revision) else: yield None def lookup_revision_message(rev_sha1_git) -> Dict[str, bytes]: """Return the raw message of the revision with sha1 revision_sha1_git. Args: revision_sha1_git: The revision's sha1 as hexadecimal Returns: Decoded revision message as dict {'message': } Raises: ValueError if the identifier provided is not of sha1 nature. NotFoundExc if the revision is not found, or if it has no message """ sha1_git_bin = _to_sha1_bin(rev_sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision with sha1_git {rev_sha1_git} not found.") if not revision.message: raise NotFoundExc(f"No message for revision with sha1_git {rev_sha1_git}.") return {"message": revision.message} def _lookup_revision_id_by(origin, branch_name, timestamp): def _get_snapshot_branch(snapshot, branch_name): snapshot = lookup_snapshot( visit["snapshot"], branches_from=branch_name, branches_count=10, branch_name_exclude_prefix=None, ) branch = None if branch_name in snapshot["branches"]: branch = snapshot["branches"][branch_name] return branch if isinstance(origin, int): origin = {"id": origin} elif isinstance(origin, str): origin = {"url": origin} else: raise TypeError('"origin" must be an int or a string.') from swh.web.common.origin_visits import get_origin_visit visit = get_origin_visit(origin, visit_ts=timestamp) branch = _get_snapshot_branch(visit["snapshot"], branch_name) rev_id = None if branch and branch["target_type"] == "revision": rev_id = branch["target"] elif branch and branch["target_type"] == "alias": branch = _get_snapshot_branch(visit["snapshot"], branch["target"]) if branch and branch["target_type"] == "revision": rev_id = branch["target"] if not rev_id: raise NotFoundExc( "Revision for origin %s and branch %s not found." % (origin.get("url"), branch_name) ) return rev_id def lookup_revision_by(origin, branch_name="HEAD", timestamp=None): """Lookup revision by origin, snapshot branch name and visit timestamp. If branch_name is not provided, lookup using 'HEAD' as default. If timestamp is not provided, use the most recent. Args: origin (Union[int,str]): origin of the revision branch_name (str): snapshot branch name timestamp (str/int): origin visit time frame Returns: dict: The revision matching the criterions Raises: NotFoundExc if no revision corresponds to the criterion """ rev_id = _lookup_revision_id_by(origin, branch_name, timestamp) return lookup_revision(rev_id) def lookup_revision_log(rev_sha1_git, limit): """Lookup revision log by revision id. Args: rev_sha1_git (str): The revision's sha1 as hexadecimal limit (int): the maximum number of revisions returned Returns: list: Revision log as list of revision dicts Raises: ValueError: if the identifier provided is not of sha1 nature. swh.web.common.exc.NotFoundExc: if there is no revision with the provided sha1_git. """ lookup_revision(rev_sha1_git) sha1_git_bin = _to_sha1_bin(rev_sha1_git) revision_entries = storage.revision_log([sha1_git_bin], limit=limit) return map(converters.from_revision, revision_entries) def lookup_revision_log_by(origin, branch_name, timestamp, limit): """Lookup revision by origin, snapshot branch name and visit timestamp. Args: origin (Union[int,str]): origin of the revision branch_name (str): snapshot branch timestamp (str/int): origin visit time frame limit (int): the maximum number of revisions returned Returns: list: Revision log as list of revision dicts Raises: swh.web.common.exc.NotFoundExc: if no revision corresponds to the criterion """ rev_id = _lookup_revision_id_by(origin, branch_name, timestamp) return lookup_revision_log(rev_id, limit) def lookup_revision_with_context_by( origin, branch_name, timestamp, sha1_git, limit=100 ): """Return information about revision sha1_git, limited to the sub-graph of all transitive parents of sha1_git_root. sha1_git_root being resolved through the lookup of a revision by origin, branch_name and ts. In other words, sha1_git is an ancestor of sha1_git_root. Args: - origin: origin of the revision. - branch_name: revision's branch. - timestamp: revision's time frame. - sha1_git: one of sha1_git_root's ancestors. - limit: limit the lookup to 100 revisions back. Returns: Pair of (root_revision, revision). Information on sha1_git if it is an ancestor of sha1_git_root including children leading to sha1_git_root Raises: - BadInputExc in case of unknown algo_hash or bad hash. - NotFoundExc if either revision is not found or if sha1_git is not an ancestor of sha1_git_root. """ rev_root_id = _lookup_revision_id_by(origin, branch_name, timestamp) rev_root_id_bin = hashutil.hash_to_bytes(rev_root_id) rev_root = storage.revision_get([rev_root_id_bin])[0] return ( converters.from_revision(rev_root) if rev_root else None, lookup_revision_with_context(rev_root, sha1_git, limit), ) def lookup_revision_with_context( sha1_git_root: Union[str, Dict[str, Any], Revision], sha1_git: str, limit: int = 100 ) -> Dict[str, Any]: """Return information about revision sha1_git, limited to the sub-graph of all transitive parents of sha1_git_root. In other words, sha1_git is an ancestor of sha1_git_root. Args: sha1_git_root: latest revision. The type is either a sha1 (as an hex string) or a non converted dict. sha1_git: one of sha1_git_root's ancestors limit: limit the lookup to 100 revisions back Returns: Information on sha1_git if it is an ancestor of sha1_git_root including children leading to sha1_git_root Raises: BadInputExc in case of unknown algo_hash or bad hash NotFoundExc if either revision is not found or if sha1_git is not an ancestor of sha1_git_root """ sha1_git_bin = _to_sha1_bin(sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision {sha1_git} not found") if isinstance(sha1_git_root, str): sha1_git_root_bin = _to_sha1_bin(sha1_git_root) revision_root = storage.revision_get([sha1_git_root_bin])[0] if not revision_root: raise NotFoundExc(f"Revision root {sha1_git_root} not found") elif isinstance(sha1_git_root, Revision): sha1_git_root_bin = sha1_git_root.id else: sha1_git_root_bin = sha1_git_root["id"] revision_log = storage.revision_log([sha1_git_root_bin], limit=limit) parents: Dict[str, List[str]] = {} children = defaultdict(list) for rev in revision_log: rev_id = rev["id"] parents[rev_id] = [] for parent_id in rev["parents"]: parents[rev_id].append(parent_id) children[parent_id].append(rev_id) if revision.id not in parents: raise NotFoundExc(f"Revision {sha1_git} is not an ancestor of {sha1_git_root}") revision_d = revision.to_dict() revision_d["children"] = children[revision.id] return converters.from_revision(revision_d) def lookup_directory_with_revision(sha1_git, dir_path=None, with_data=False): """Return information on directory pointed by revision with sha1_git. If dir_path is not provided, display top level directory. Otherwise, display the directory pointed by dir_path (if it exists). Args: sha1_git: revision's hash. dir_path: optional directory pointed to by that revision. with_data: boolean that indicates to retrieve the raw data if the path resolves to a content. Default to False (for the api) Returns: Information on the directory pointed to by that revision. Raises: BadInputExc in case of unknown algo_hash or bad hash. NotFoundExc either if the revision is not found or the path referenced does not exist. NotImplementedError in case of dir_path exists but do not reference a type 'dir' or 'file'. """ sha1_git_bin = _to_sha1_bin(sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision {sha1_git} not found") dir_sha1_git_bin = revision.directory if dir_path: paths = dir_path.strip(os.path.sep).split(os.path.sep) entity = storage.directory_entry_get_by_path( dir_sha1_git_bin, list(map(lambda p: p.encode("utf-8"), paths)) ) if not entity: raise NotFoundExc( "Directory or File '%s' pointed to by revision %s not found" % (dir_path, sha1_git) ) else: entity = {"type": "dir", "target": dir_sha1_git_bin} if entity["type"] == "dir": directory_entries = storage.directory_ls(entity["target"]) or [] return { "type": "dir", "path": "." if not dir_path else dir_path, "revision": sha1_git, "content": list(map(converters.from_directory_entry, directory_entries)), } elif entity["type"] == "file": # content content = _first_element(storage.content_find({"sha1_git": entity["target"]})) if not content: raise NotFoundExc(f"Content not found for revision {sha1_git}") content_d = content.to_dict() if with_data: data = storage.content_get_data(content.sha1) if data: content_d["data"] = data return { "type": "file", "path": "." if not dir_path else dir_path, "revision": sha1_git, "content": converters.from_content(content_d), } elif entity["type"] == "rev": # revision revision = storage.revision_get([entity["target"]])[0] return { "type": "rev", "path": "." if not dir_path else dir_path, "revision": sha1_git, "content": converters.from_revision(revision) if revision else None, } else: raise NotImplementedError("Entity of type %s not implemented." % entity["type"]) def lookup_content(q: str) -> Dict[str, Any]: """Lookup the content designed by q. Args: q: The release's sha1 as hexadecimal Raises: NotFoundExc if the requested content is not found """ algo, hash_ = query.parse_hash(q) c = _first_element(storage.content_find({algo: hash_})) if not c: hhex = hashutil.hash_to_hex(hash_) raise NotFoundExc(f"Content with {algo} checksum equals to {hhex} not found!") return converters.from_content(c.to_dict()) def lookup_content_raw(q: str) -> Dict[str, Any]: """Lookup the content defined by q. Args: q: query string of the form Returns: dict with 'sha1' and 'data' keys. data representing its raw data decoded. Raises: NotFoundExc if the requested content is not found or if the content bytes are not available in the storage """ c = lookup_content(q) content_sha1_bytes = hashutil.hash_to_bytes(c["checksums"]["sha1"]) content_data = storage.content_get_data(content_sha1_bytes) if content_data is None: algo, hash_ = query.parse_hash(q) raise NotFoundExc( f"Bytes of content with {algo} checksum equals " f"to {hashutil.hash_to_hex(hash_)} are not available!" ) return converters.from_content({"sha1": content_sha1_bytes, "data": content_data}) def stat_counters(): """Return the stat counters for Software Heritage Returns: A dict mapping textual labels to integer values. """ res = {} if counters and config.get_config()["counters_backend"] == "swh-counters": res = counters.get_counts( ["origin", "revision", "content", "directory", "release", "person"] ) else: res = storage.stat_counters() return res def _lookup_origin_visits( origin_url: str, last_visit: Optional[int] = None, limit: int = 10 ) -> Iterator[OriginVisit]: """Yields the origin origins' visits. Args: origin_url (str): origin to list visits for last_visit (int): last visit to lookup from limit (int): Number of elements max to display Yields: OriginVisit for that origin """ limit = min(limit, MAX_LIMIT) page_token: Optional[str] if last_visit is not None: page_token = str(last_visit) else: page_token = None visit_page = storage.origin_visit_get( origin_url, page_token=page_token, limit=limit ) yield from visit_page.results def lookup_origin_visits( origin: str, last_visit: Optional[int] = None, per_page: int = 10 ) -> Iterator[OriginVisitInfo]: """Yields the origin origins' visits. Args: origin: origin to list visits for Yields: Dictionaries of origin_visit for that origin """ for visit in _lookup_origin_visits(origin, last_visit=last_visit, limit=per_page): visit_status = storage.origin_visit_status_get_latest(origin, visit.visit) yield converters.from_origin_visit( {**visit_status.to_dict(), "type": visit.type} ) def lookup_origin_visit_latest( origin_url: str, require_snapshot: bool = False, type: Optional[str] = None, allowed_statuses: Optional[List[str]] = None, ) -> Optional[OriginVisitInfo]: """Return the origin's latest visit Args: origin_url: origin to list visits for type: Optional visit type to filter on (e.g git, tar, dsc, svn, hg, npm, pypi, ...) allowed_statuses: list of visit statuses considered to find the latest visit. For instance, ``allowed_statuses=['full']`` will only consider visits that have successfully run to completion. require_snapshot: filter out origins without a snapshot Returns: The origin visit info as dict if found """ visit_status = origin_get_latest_visit_status( storage, origin_url, type=type, allowed_statuses=allowed_statuses, require_snapshot=require_snapshot, ) return ( converters.from_origin_visit(visit_status.to_dict()) if visit_status else None ) def lookup_origin_visit(origin_url: str, visit_id: int) -> OriginVisitInfo: """Return information about visit visit_id with origin origin. Args: origin: origin concerned by the visit visit_id: the visit identifier to lookup Raises: NotFoundExc if no origin visit matching the criteria is found Returns: The dict origin_visit concerned """ visit = storage.origin_visit_get_by(origin_url, visit_id) visit_status = storage.origin_visit_status_get_latest(origin_url, visit_id) if not visit: raise NotFoundExc( f"Origin {origin_url} or its visit with id {visit_id} not found!" ) return converters.from_origin_visit({**visit_status.to_dict(), "type": visit.type}) def origin_visit_find_by_date( - origin_url: str, visit_date: datetime.datetime + origin_url: str, visit_date: datetime.datetime, greater_or_equal: bool = True ) -> Optional[OriginVisitInfo]: """Retrieve origin visit status whose date is most recent than the provided visit_date. Args: origin_url: origin concerned by the visit visit_date: provided visit date + greater_or_equal: ensure returned visit has a date greater or equal + than the one passed as parameter Returns: The dict origin_visit_status matching the criteria if any. """ visit = storage.origin_visit_find_by_date(origin_url, visit_date) - if visit and visit.date < visit_date: + if greater_or_equal and visit and visit.date < visit_date: # when visit is anterior to the provided date, trying to find another one most # recent visits = storage.origin_visit_get( origin_url, page_token=str(visit.visit), limit=1, ).results visit = visits[0] if visits else None if not visit: return None visit_status = storage.origin_visit_status_get_latest(origin_url, visit.visit) return converters.from_origin_visit({**visit_status.to_dict(), "type": visit.type}) def lookup_snapshot_sizes( snapshot_id: str, branch_name_exclude_prefix: Optional[str] = "refs/pull/" ) -> Dict[str, int]: """Count the number of branches in the snapshot with the given id. Args: snapshot_id (str): sha1 identifier of the snapshot Returns: dict: A dict whose keys are the target types of branches and values their corresponding amount """ snapshot_id_bin = _to_sha1_bin(snapshot_id) snapshot_sizes = dict.fromkeys(("alias", "release", "revision"), 0) branch_counts = storage.snapshot_count_branches( snapshot_id_bin, branch_name_exclude_prefix.encode() if branch_name_exclude_prefix else None, ) # remove possible None key returned by snapshot_count_branches # when null branches are present in the snapshot branch_counts.pop(None, None) snapshot_sizes.update(branch_counts) return snapshot_sizes def lookup_snapshot( snapshot_id: str, branches_from: str = "", branches_count: int = 1000, target_types: Optional[List[str]] = None, branch_name_include_substring: Optional[str] = None, branch_name_exclude_prefix: Optional[str] = "refs/pull/", ) -> Dict[str, Any]: """Return information about a snapshot, aka the list of named branches found during a specific visit of an origin. Args: snapshot_id: sha1 identifier of the snapshot branches_from: optional parameter used to skip branches whose name is lesser than it before returning them branches_count: optional parameter used to restrain the amount of returned branches target_types: optional 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', 'alias'`) branch_name_include_substring: if provided, only return branches whose name contains given substring branch_name_exclude_prefix: if provided, do not return branches whose name starts with given pattern Raises: NotFoundExc if the given snapshot_id is missing Returns: A dict filled with the snapshot content. """ snapshot_id_bin = _to_sha1_bin(snapshot_id) if storage.snapshot_missing([snapshot_id_bin]): raise NotFoundExc(f"Snapshot with id {snapshot_id} not found!") partial_branches = storage.snapshot_get_branches( snapshot_id_bin, branches_from.encode(), branches_count, target_types, branch_name_include_substring.encode() if branch_name_include_substring else None, branch_name_exclude_prefix.encode() if branch_name_exclude_prefix else None, ) return ( converters.from_partial_branches(partial_branches) if partial_branches else None ) def lookup_latest_origin_snapshot( origin: str, allowed_statuses: List[str] = None ) -> Optional[Dict[str, Any]]: """Return information about the latest snapshot of an origin. .. warning:: At most 1000 branches contained in the snapshot will be returned for performance reasons. Args: origin: URL or integer identifier of the origin allowed_statuses: list of visit statuses considered to find the latest snapshot for the visit. For instance, ``allowed_statuses=['full']`` will only consider visits that have successfully run to completion. Returns: A dict filled with the snapshot content. """ snp = snapshot_get_latest( storage, origin, allowed_statuses=allowed_statuses, branches_count=1000 ) return converters.from_snapshot(snp.to_dict()) if snp is not None else None def lookup_snapshot_alias( snapshot_id: str, alias_name: str ) -> Optional[Dict[str, Any]]: """Try to resolve a branch alias in a snapshot. Args: snapshot_id: hexadecimal representation of a snapshot id alias_name: name of the branch alias to resolve Returns: Target branch information or None if the alias does not exist or target a dangling branch. """ resolved_alias = snapshot_resolve_alias( storage, _to_sha1_bin(snapshot_id), alias_name.encode() ) return ( converters.from_swh(resolved_alias.to_dict(), hashess={"target"}) if resolved_alias is not None else None ) def lookup_revision_through(revision, limit=100): """Retrieve a revision from the criterion stored in revision dictionary. Args: revision: Dictionary of criterion to lookup the revision with. Here are the supported combination of possible values: - origin_url, branch_name, ts, sha1_git - origin_url, branch_name, ts - sha1_git_root, sha1_git - sha1_git Returns: None if the revision is not found or the actual revision. """ if ( "origin_url" in revision and "branch_name" in revision and "ts" in revision and "sha1_git" in revision ): return lookup_revision_with_context_by( revision["origin_url"], revision["branch_name"], revision["ts"], revision["sha1_git"], limit, ) if "origin_url" in revision and "branch_name" in revision and "ts" in revision: return lookup_revision_by( revision["origin_url"], revision["branch_name"], revision["ts"] ) if "sha1_git_root" in revision and "sha1_git" in revision: return lookup_revision_with_context( revision["sha1_git_root"], revision["sha1_git"], limit ) if "sha1_git" in revision: return lookup_revision(revision["sha1_git"]) # this should not happen raise NotImplementedError("Should not happen!") def lookup_directory_through_revision(revision, path=None, limit=100, with_data=False): """Retrieve the directory information from the revision. Args: revision: dictionary of criterion representing a revision to lookup path: directory's path to lookup. limit: optional query parameter to limit the revisions log (default to 100). For now, note that this limit could impede the transitivity conclusion about sha1_git not being an ancestor of. with_data: indicate to retrieve the content's raw data if path resolves to a content. Returns: The directory pointing to by the revision criterions at path. """ rev = lookup_revision_through(revision, limit) if not rev: raise NotFoundExc("Revision with criterion %s not found!" % revision) return (rev["id"], lookup_directory_with_revision(rev["id"], path, with_data)) def _vault_request(vault_fn, bundle_type: str, swhid: CoreSWHID, **kwargs): try: return vault_fn(bundle_type, swhid, **kwargs) except VaultNotFoundExc: return None def vault_cook(bundle_type: str, swhid: CoreSWHID, email=None): """Cook a vault bundle. """ return _vault_request(vault.cook, bundle_type, swhid, email=email) def vault_fetch(bundle_type: str, swhid: CoreSWHID): """Fetch a vault bundle. """ return _vault_request(vault.fetch, bundle_type, swhid) def vault_progress(bundle_type: str, swhid: CoreSWHID): """Get the current progress of a vault bundle. """ return _vault_request(vault.progress, bundle_type, swhid) def diff_revision(rev_id): """Get the list of file changes (insertion / deletion / modification / renaming) for a particular revision. """ rev_sha1_git_bin = _to_sha1_bin(rev_id) changes = diff.diff_revision(storage, rev_sha1_git_bin, track_renaming=True) for change in changes: change["from"] = converters.from_directory_entry(change["from"]) change["to"] = converters.from_directory_entry(change["to"]) if change["from_path"]: change["from_path"] = change["from_path"].decode("utf-8") if change["to_path"]: change["to_path"] = change["to_path"].decode("utf-8") return changes class _RevisionsWalkerProxy(object): """ Proxy class wrapping a revisions walker iterator from swh-storage and performing needed conversions. """ def __init__(self, rev_walker_type, rev_start, *args, **kwargs): rev_start_bin = hashutil.hash_to_bytes(rev_start) self.revisions_walker = revisions_walker.get_revisions_walker( rev_walker_type, storage, rev_start_bin, *args, **kwargs ) def export_state(self): return self.revisions_walker.export_state() def __next__(self): return converters.from_revision(next(self.revisions_walker)) def __iter__(self): return self def get_revisions_walker(rev_walker_type, rev_start, *args, **kwargs): """ Utility function to instantiate a revisions walker of a given type, see :mod:`swh.storage.algos.revisions_walker`. Args: rev_walker_type (str): the type of revisions walker to return, possible values are: ``committer_date``, ``dfs``, ``dfs_post``, ``bfs`` and ``path`` rev_start (str): hexadecimal representation of a revision identifier args (list): position arguments to pass to the revisions walker constructor kwargs (dict): keyword arguments to pass to the revisions walker constructor """ # first check if the provided revision is valid lookup_revision(rev_start) return _RevisionsWalkerProxy(rev_walker_type, rev_start, *args, **kwargs) def lookup_object(object_type: ObjectType, object_id: str) -> Dict[str, Any]: """ Utility function for looking up an object in the archive by its type and id. Args: object_type (str): the type of object to lookup, either *content*, *directory*, *release*, *revision* or *snapshot* object_id (str): the *sha1_git* checksum identifier in hexadecimal form of the object to lookup Returns: Dict[str, Any]: A dictionary describing the object or a list of dictionary for the directory object type. Raises: swh.web.common.exc.NotFoundExc: if the object could not be found in the archive BadInputExc: if the object identifier is invalid """ if object_type == ObjectType.CONTENT: return lookup_content(f"sha1_git:{object_id}") elif object_type == ObjectType.DIRECTORY: return {"id": object_id, "content": list(lookup_directory(object_id))} elif object_type == ObjectType.RELEASE: return lookup_release(object_id) elif object_type == ObjectType.REVISION: return lookup_revision(object_id) elif object_type == ObjectType.SNAPSHOT: return lookup_snapshot(object_id) else: raise ValueError(f"Unexpected object type variant: {object_type}") def lookup_missing_hashes(grouped_swhids: Dict[str, List[bytes]]) -> Set[str]: """Lookup missing Software Heritage persistent identifier hash, using batch processing. Args: A dictionary with: keys: object types values: object hashes Returns: A set(hexadecimal) of the hashes not found in the storage """ missing_hashes = [] for obj_type, obj_ids in grouped_swhids.items(): if obj_type == ObjectType.CONTENT: missing_hashes.append(storage.content_missing_per_sha1_git(obj_ids)) elif obj_type == ObjectType.DIRECTORY: missing_hashes.append(storage.directory_missing(obj_ids)) elif obj_type == ObjectType.REVISION: missing_hashes.append(storage.revision_missing(obj_ids)) elif obj_type == ObjectType.RELEASE: missing_hashes.append(storage.release_missing(obj_ids)) elif obj_type == ObjectType.SNAPSHOT: missing_hashes.append(storage.snapshot_missing(obj_ids)) missing = set( map(lambda x: hashutil.hash_to_hex(x), itertools.chain(*missing_hashes)) ) return missing def lookup_origins_by_sha1s(sha1s: List[str]) -> Iterator[Optional[OriginInfo]]: """Lookup origins from the sha1 hash values of their URLs. Args: sha1s: list of sha1s hexadecimal representation Yields: origin information as dict """ sha1s_bytes = [hashutil.hash_to_bytes(sha1) for sha1 in sha1s] origins = storage.origin_get_by_sha1(sha1s_bytes) for origin in origins: yield converters.from_origin(origin) diff --git a/swh/web/common/origin_visits.py b/swh/web/common/origin_visits.py index 24cb5ff0..a7412563 100644 --- a/swh/web/common/origin_visits.py +++ b/swh/web/common/origin_visits.py @@ -1,199 +1,175 @@ -# Copyright (C) 2018-2019 The Software Heritage developers +# Copyright (C) 2018-2022 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 math from typing import List, Optional from django.core.cache import cache from swh.web.common import archive from swh.web.common.exc import NotFoundExc from swh.web.common.typing import OriginInfo, OriginVisitInfo from swh.web.common.utils import parse_iso8601_date_to_utc def get_origin_visits(origin_info: OriginInfo) -> List[OriginVisitInfo]: """Function that returns the list of visits for a swh origin. That list is put in cache in order to speedup the navigation in the swh web browse ui. The returned visits are sorted according to their date in ascending order. Args: origin_info: dict describing the origin to fetch visits from Returns: A list of dict describing the origin visits Raises: swh.web.common.exc.NotFoundExc: if the origin is not found """ from swh.web.common import archive if "url" in origin_info: origin_url = origin_info["url"] else: origin_url = archive.lookup_origin(origin_info)["url"] cache_entry_id = "origin_visits_%s" % origin_url cache_entry = cache.get(cache_entry_id) last_visit = 0 origin_visits = [] new_visits = [] per_page = archive.MAX_LIMIT if cache_entry: origin_visits = cache_entry last_visit = cache_entry[-1]["visit"] new_visits = list( archive.lookup_origin_visits( origin_url, last_visit=last_visit, per_page=per_page ) ) last_visit += len(new_visits) if not new_visits: last_snp = archive.lookup_latest_origin_snapshot(origin_url) if not last_snp or last_snp["id"] == cache_entry[-1]["snapshot"]: return cache_entry # get new visits that we did not retrieve yet while 1: visits = list( archive.lookup_origin_visits( origin_url, last_visit=last_visit, per_page=per_page ) ) new_visits += visits if len(visits) < per_page: break last_visit += per_page def _visit_sort_key(visit): ts = parse_iso8601_date_to_utc(visit["date"]).timestamp() return ts + (float(visit["visit"]) / 10e3) # cache entry is already sorted with oldest visits origin_visits += sorted(new_visits, key=lambda v: _visit_sort_key(v)) cache.set(cache_entry_id, origin_visits) return origin_visits def get_origin_visit( origin_info: OriginInfo, visit_ts: Optional[str] = None, visit_id: Optional[int] = None, snapshot_id: Optional[str] = None, ) -> OriginVisitInfo: """Function that returns information about a visit for a given origin. If a timestamp is provided, the closest visit from that timestamp is returned. If a snapshot identifier is provided, the first visit with that snapshot is returned. If no search hints are provided, return the most recent full visit with a valid snapshot or the most recent partial visit with a valid snapshot otherwise. Args: origin_info: a dict filled with origin information visit_ts: an ISO 8601 datetime string to parse snapshot_id: a snapshot identifier Returns: A dict containing the visit info. Raises: swh.web.common.exc.NotFoundExc: if no visit can be found """ # returns the latest full visit with a valid snapshot visit = archive.lookup_origin_visit_latest( origin_info["url"], allowed_statuses=["full"], require_snapshot=True ) if not visit: # or the latest partial visit with a valid snapshot otherwise visit = archive.lookup_origin_visit_latest( origin_info["url"], allowed_statuses=["partial"], require_snapshot=True ) if not visit_ts and not visit_id and not snapshot_id: if visit: return visit else: raise NotFoundExc( f"No valid visit for origin with url {origin_info['url']} found!" ) # no need to fetch all visits list and search in it if the latest # visit matches some criteria if visit and (visit["snapshot"] == snapshot_id or visit["visit"] == visit_id): return visit + if visit_id: + return archive.lookup_origin_visit(origin_info["url"], visit_id) + + if visit_ts: + visit = archive.origin_visit_find_by_date( + origin_info["url"], + parse_iso8601_date_to_utc(visit_ts), + greater_or_equal=False, + ) + if visit is not None: + return visit + else: + raise NotFoundExc( + ( + "Visit with timestamp %s for origin with " + "url %s not found!" % (visit_ts, origin_info["url"]) + ) + ) + visits = get_origin_visits(origin_info) if not visits: raise NotFoundExc( f"No visits associated to origin with url {origin_info['url']}!" ) if snapshot_id: visits = [v for v in visits if v["snapshot"] == snapshot_id] if len(visits) == 0: raise NotFoundExc( ( "Visit for snapshot with id %s for origin with" " url %s not found!" % (snapshot_id, origin_info["url"]) ) ) return visits[0] - if visit_id: - visits = [v for v in visits if v["visit"] == int(visit_id)] - if len(visits) == 0: - raise NotFoundExc( - ( - "Visit with id %s for origin with" - " url %s not found!" % (visit_id, origin_info["url"]) - ) - ) - return visits[0] - - if visit_ts: - - target_visit_ts = math.floor(parse_iso8601_date_to_utc(visit_ts).timestamp()) - - # Find the visit with date closest to the target (in absolute value) - (abs_time_delta, visit_idx) = min( - ( - (math.floor(parse_iso8601_date_to_utc(visit["date"]).timestamp()), i) - for (i, visit) in enumerate(visits) - ), - key=lambda ts_and_i: abs(ts_and_i[0] - target_visit_ts), - ) - - if visit_idx is not None: - visit = visits[visit_idx] - # If multiple visits have the same date, select the one with - # the largest id. - 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 " - "url %s not found!" % (visit_ts, origin_info["url"]) - ) - ) return visits[-1] diff --git a/swh/web/tests/browse/views/test_origin.py b/swh/web/tests/browse/views/test_origin.py index 1c48b508..a0ab9ec9 100644 --- a/swh/web/tests/browse/views/test_origin.py +++ b/swh/web/tests/browse/views/test_origin.py @@ -1,893 +1,888 @@ -# Copyright (C) 2017-2021 The Software Heritage developers +# Copyright (C) 2017-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import random import re from hypothesis import given import pytest from django.utils.html import escape from swh.model.hashutil import hash_to_bytes from swh.model.model import ( OriginVisit, OriginVisitStatus, Snapshot, SnapshotBranch, TargetType, ) from swh.model.swhids import ObjectType from swh.storage.utils import now from swh.web.browse.snapshot_context import process_snapshot_branches from swh.web.common.exc import NotFoundExc from swh.web.common.identifiers import gen_swhid from swh.web.common.utils import format_utc_iso_date, parse_iso8601_date_to_utc, reverse from swh.web.tests.django_asserts import assert_contains, assert_not_contains from swh.web.tests.strategies import new_origin, new_snapshot, visit_dates from swh.web.tests.utils import check_html_get_response def test_origin_visits_browse(client, archive_data, origin_with_multiple_visits): origin_url = origin_with_multiple_visits["url"] url = reverse("browse-origin-visits", query_params={"origin_url": origin_url}) resp = check_html_get_response( client, url, status_code=200, template_used="browse/origin-visits.html" ) visits = archive_data.origin_visit_get(origin_url) for v in visits: vdate = format_utc_iso_date(v["date"], "%Y-%m-%dT%H:%M:%SZ") browse_dir_url = reverse( "browse-origin-directory", query_params={"origin_url": origin_url, "timestamp": vdate}, ) assert_contains(resp, browse_dir_url) _check_origin_link(resp, origin_url) def test_origin_root_directory_view(client, archive_data, swh_scheduler, origin): origin_visits = archive_data.origin_visit_get(origin["url"]) visit = origin_visits[-1] snapshot = archive_data.snapshot_get(visit["snapshot"]) snapshot_sizes = archive_data.snapshot_count_branches(snapshot["id"]) head_rev_id = archive_data.snapshot_get_head(snapshot) head_rev = archive_data.revision_get(head_rev_id) root_dir_sha1 = head_rev["directory"] dir_content = archive_data.directory_ls(root_dir_sha1) branches, releases, _ = process_snapshot_branches(snapshot) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, dir_content, ) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, dir_content, visit_id=visit["visit"], ) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, dir_content, timestamp=visit["date"], ) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, dir_content, snapshot_id=visit["snapshot"], ) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, dir_content, ) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, dir_content, visit_id=visit["visit"], ) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, dir_content, timestamp=visit["date"], ) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, dir_content, snapshot_id=visit["snapshot"], ) def test_origin_sub_directory_view(client, archive_data, swh_scheduler, origin): origin_visits = archive_data.origin_visit_get(origin["url"]) visit = origin_visits[-1] snapshot = archive_data.snapshot_get(visit["snapshot"]) snapshot_sizes = archive_data.snapshot_count_branches(snapshot["id"]) head_rev_id = archive_data.snapshot_get_head(snapshot) head_rev = archive_data.revision_get(head_rev_id) root_dir_sha1 = head_rev["directory"] subdirs = [ e for e in archive_data.directory_ls(root_dir_sha1) if e["type"] == "dir" ] branches, releases, _ = process_snapshot_branches(snapshot) if len(subdirs) == 0: return subdir = random.choice(subdirs) subdir_content = archive_data.directory_ls(subdir["target"]) subdir_path = subdir["name"] _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, subdir_content, path=subdir_path, ) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, subdir_content, path=subdir_path, visit_id=visit["visit"], ) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, subdir_content, path=subdir_path, timestamp=visit["date"], ) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, subdir_content, path=subdir_path, snapshot_id=visit["snapshot"], ) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, subdir_content, path=subdir_path, ) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, subdir_content, path=subdir_path, visit_id=visit["visit"], ) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, subdir_content, path=subdir_path, timestamp=visit["date"], ) _origin_directory_view_test_helper( client, archive_data, origin, visit, snapshot_sizes, branches, releases, root_dir_sha1, subdir_content, path=subdir_path, snapshot_id=visit["snapshot"], ) @given( new_origin(), new_snapshot(min_size=4, max_size=4), visit_dates(), ) def test_origin_snapshot_null_branch( client, archive_data, revisions_list, new_origin, new_snapshot, visit_dates, ): revisions = revisions_list(size=4) snp_dict = new_snapshot.to_dict() archive_data.origin_add([new_origin]) for i, branch in enumerate(snp_dict["branches"].keys()): if i == 0: snp_dict["branches"][branch] = None else: snp_dict["branches"][branch] = { "target_type": "revision", "target": hash_to_bytes(revisions[i - 1]), } archive_data.snapshot_add([Snapshot.from_dict(snp_dict)]) visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=visit_dates[0], type="git",)] )[0] visit_status = OriginVisitStatus( origin=new_origin.url, visit=visit.visit, date=now(), status="partial", snapshot=snp_dict["id"], ) archive_data.origin_visit_status_add([visit_status]) url = reverse( "browse-origin-directory", query_params={"origin_url": new_origin.url} ) check_html_get_response( client, url, status_code=200, template_used="browse/directory.html" ) @given( new_origin(), new_snapshot(min_size=4, max_size=4), visit_dates(), ) def test_origin_snapshot_invalid_branch( client, archive_data, revisions_list, new_origin, new_snapshot, visit_dates, ): revisions = revisions_list(size=4) snp_dict = new_snapshot.to_dict() archive_data.origin_add([new_origin]) for i, branch in enumerate(snp_dict["branches"].keys()): snp_dict["branches"][branch] = { "target_type": "revision", "target": hash_to_bytes(revisions[i]), } archive_data.snapshot_add([Snapshot.from_dict(snp_dict)]) visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=visit_dates[0], type="git",)] )[0] visit_status = OriginVisitStatus( origin=new_origin.url, visit=visit.visit, date=now(), status="full", snapshot=snp_dict["id"], ) archive_data.origin_visit_status_add([visit_status]) url = reverse( "browse-origin-directory", query_params={"origin_url": new_origin.url, "branch": "invalid_branch"}, ) check_html_get_response(client, url, status_code=404, template_used="error.html") @given(new_origin()) def test_browse_visits_origin_not_found(client, new_origin): url = reverse("browse-origin-visits", query_params={"origin_url": new_origin.url}) resp = check_html_get_response( client, url, status_code=404, template_used="error.html" ) assert_contains( resp, f"Origin with url {new_origin.url} not found", status_code=404 ) def test_browse_origin_directory_no_visit(client, mocker, origin): mock_get_origin_visits = mocker.patch( "swh.web.common.origin_visits.get_origin_visits" ) mock_get_origin_visits.return_value = [] mock_archive = mocker.patch("swh.web.common.origin_visits.archive") mock_archive.lookup_origin_visit_latest.return_value = None url = reverse("browse-origin-directory", query_params={"origin_url": origin["url"]}) resp = check_html_get_response( client, url, status_code=404, template_used="error.html" ) assert_contains(resp, "No valid visit", status_code=404) assert not mock_get_origin_visits.called -def test_browse_origin_directory_unknown_visit(client, mocker, origin): - mock_get_origin_visits = mocker.patch( - "swh.web.common.origin_visits.get_origin_visits" - ) - mock_get_origin_visits.return_value = [{"visit": 1}] +def test_browse_origin_directory_unknown_visit(client, origin): url = reverse( "browse-origin-directory", - query_params={"origin_url": origin["url"], "visit_id": 2}, + query_params={"origin_url": origin["url"], "visit_id": 200}, ) resp = check_html_get_response( client, url, status_code=404, template_used="error.html" ) - assert re.search("Visit.*not found", resp.content.decode("utf-8")) - assert mock_get_origin_visits.called + assert re.search("visit.*not found", resp.content.decode("utf-8")) def test_browse_origin_directory_not_found(client, origin): url = reverse( "browse-origin-directory", query_params={"origin_url": origin["url"], "path": "/invalid/dir/path/"}, ) resp = check_html_get_response( client, url, status_code=404, template_used="browse/directory.html" ) assert re.search("Directory.*not found", resp.content.decode("utf-8")) def _add_empty_snapshot_origin(new_origin, archive_data): snapshot = Snapshot(branches={}) archive_data.origin_add([new_origin]) archive_data.snapshot_add([snapshot]) visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=now(), type="git",)] )[0] visit_status = OriginVisitStatus( origin=new_origin.url, visit=visit.visit, date=now(), status="full", snapshot=snapshot.id, ) archive_data.origin_visit_status_add([visit_status]) @pytest.mark.django_db @pytest.mark.parametrize("object_type", ["directory"]) @given(new_origin()) def test_browse_origin_content_directory_empty_snapshot( client, staff_user, archive_data, object_type, new_origin ): _add_empty_snapshot_origin(new_origin, archive_data) # to check proper generation of raw extrinsic metadata api links client.force_login(staff_user) url = reverse( f"browse-origin-{object_type}", query_params={"origin_url": new_origin.url, "path": "baz"}, ) resp = check_html_get_response( client, url, status_code=200, template_used=f"browse/{object_type}.html" ) assert re.search("snapshot.*is empty", resp.content.decode("utf-8")) def test_browse_directory_snapshot_not_found(client, mocker, origin): mock_get_snapshot_context = mocker.patch( "swh.web.browse.snapshot_context.get_snapshot_context" ) mock_get_snapshot_context.side_effect = NotFoundExc("Snapshot not found") url = reverse("browse-origin-directory", query_params={"origin_url": origin["url"]}) resp = check_html_get_response( client, url, status_code=404, template_used="error.html" ) assert_contains(resp, "Snapshot not found", status_code=404) assert mock_get_snapshot_context.called @given(new_origin()) def test_origin_empty_snapshot(client, archive_data, new_origin): _add_empty_snapshot_origin(new_origin, archive_data) url = reverse( "browse-origin-directory", query_params={"origin_url": new_origin.url} ) resp = check_html_get_response( client, url, status_code=200, template_used="browse/directory.html" ) resp_content = resp.content.decode("utf-8") assert re.search("snapshot.*is empty", resp_content) assert not re.search("swh-tr-link", resp_content) @given(new_origin()) def test_origin_empty_snapshot_null_revision(client, archive_data, new_origin): snapshot = Snapshot( branches={ b"HEAD": SnapshotBranch( target="refs/head/master".encode(), target_type=TargetType.ALIAS, ), b"refs/head/master": None, } ) archive_data.origin_add([new_origin]) archive_data.snapshot_add([snapshot]) visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=now(), type="git",)] )[0] visit_status = OriginVisitStatus( origin=new_origin.url, visit=visit.visit, date=now(), status="partial", snapshot=snapshot.id, ) archive_data.origin_visit_status_add([visit_status]) url = reverse( "browse-origin-directory", query_params={"origin_url": new_origin.url}, ) resp = check_html_get_response( client, url, status_code=200, template_used="browse/directory.html" ) resp_content = resp.content.decode("utf-8") assert re.search("snapshot.*is empty", resp_content) assert not re.search("swh-tr-link", resp_content) def test_origin_release_browse(client, archive_data, origin_with_releases): origin_url = origin_with_releases["url"] snapshot = archive_data.snapshot_get_latest(origin_url) release = [ b for b in snapshot["branches"].values() if b["target_type"] == "release" ][-1] release_data = archive_data.release_get(release["target"]) revision_data = archive_data.revision_get(release_data["target"]) url = reverse( "browse-origin-directory", query_params={"origin_url": origin_url, "release": release_data["name"]}, ) resp = check_html_get_response( client, url, status_code=200, template_used="browse/directory.html" ) assert_contains(resp, release_data["name"]) assert_contains(resp, release["target"]) swhid_context = { "origin": origin_url, "visit": gen_swhid(ObjectType.SNAPSHOT, snapshot["id"]), "anchor": gen_swhid(ObjectType.RELEASE, release_data["id"]), } swh_dir_id = gen_swhid( ObjectType.DIRECTORY, revision_data["directory"], metadata=swhid_context ) swh_dir_id_url = reverse("browse-swhid", url_args={"swhid": swh_dir_id}) assert_contains(resp, swh_dir_id) assert_contains(resp, swh_dir_id_url) def test_origin_release_browse_not_found(client, origin_with_releases): invalid_release_name = "swh-foo-bar" url = reverse( "browse-origin-directory", query_params={ "origin_url": origin_with_releases["url"], "release": invalid_release_name, }, ) resp = check_html_get_response( client, url, status_code=404, template_used="error.html" ) assert re.search( f"Release {invalid_release_name}.*not found", resp.content.decode("utf-8") ) @given(new_origin()) def test_origin_browse_directory_branch_with_non_resolvable_revision( client, archive_data, unknown_revision, new_origin, ): branch_name = "master" snapshot = Snapshot( branches={ branch_name.encode(): SnapshotBranch( target=hash_to_bytes(unknown_revision), target_type=TargetType.REVISION, ) } ) archive_data.origin_add([new_origin]) archive_data.snapshot_add([snapshot]) visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=now(), type="git",)] )[0] visit_status = OriginVisitStatus( origin=new_origin.url, visit=visit.visit, date=now(), status="partial", snapshot=snapshot.id, ) archive_data.origin_visit_status_add([visit_status]) url = reverse( "browse-origin-directory", query_params={"origin_url": new_origin.url, "branch": branch_name}, ) resp = check_html_get_response( client, url, status_code=200, template_used="browse/directory.html" ) assert_contains( resp, f"Revision {unknown_revision } could not be found in the archive." ) # no revision card assert_not_contains(resp, "swh-tip-revision") # no Download dropdown assert_not_contains(resp, "swh-vault-download") # no History link assert_not_contains(resp, "swh-tr-link") # no SWHIDs for directory and revision assert_not_contains(resp, "swh:1:dir:") assert_not_contains(resp, "swh:1:rev:") def test_origin_views_no_url_query_parameter(client): for browse_context in ( "directory", "visits", ): url = reverse(f"browse-origin-{browse_context}") resp = check_html_get_response( client, url, status_code=400, template_used="error.html" ) assert_contains( resp, "An origin URL must be provided as query parameter.", status_code=400, ) @given(new_origin()) @pytest.mark.parametrize("browse_context", ["log", "branches", "releases"]) def test_origin_view_redirects(client, browse_context, new_origin): query_params = {"origin_url": new_origin.url} url = reverse(f"browse-origin-{browse_context}", query_params=query_params) resp = check_html_get_response(client, url, status_code=301) assert resp["location"] == reverse( f"browse-snapshot-{browse_context}", query_params=query_params ) @given(new_origin()) @pytest.mark.parametrize("browse_context", ["content"]) def test_origin_content_view_redirects(client, browse_context, new_origin): query_params = {"origin_url": new_origin.url, "path": "test.txt"} url = reverse(f"browse-origin-{browse_context}", query_params=query_params) resp = check_html_get_response(client, url, status_code=301) assert resp["location"] == reverse( f"browse-{browse_context}", query_params=query_params ) @given(new_origin()) @pytest.mark.parametrize("browse_context", ["log", "branches", "releases"]) def test_origin_view_legacy_redirects(client, browse_context, new_origin): # Each legacy route corresponds to two URL patterns, testing both url_args = [ {"origin_url": new_origin.url}, {"origin_url": new_origin.url, "timestamp": "2021-01-23T22:24:10Z"}, ] params = {"extra-param1": "extra-param1", "extra-param2": "extra-param2"} for each_arg in url_args: url = reverse( f"browse-origin-{browse_context}-legacy", url_args=each_arg, query_params=params, ) resp = check_html_get_response(client, url, status_code=301) assert resp["location"] == reverse( f"browse-snapshot-{browse_context}", query_params={**each_arg, **params} ) @given(new_origin()) def test_origin_content_view_legacy_redirects(client, new_origin): url_args = [ {"origin_url": new_origin.url}, { "origin_url": new_origin.url, "path": "test.txt", "timestamp": "2021-01-23T22:24:10Z", }, {"origin_url": new_origin.url, "path": "test.txt"}, ] params = {"extra-param1": "extra-param1", "extra-param2": "extra-param2"} for each_arg in url_args: url = reverse( "browse-origin-content-legacy", url_args=each_arg, query_params=params, ) resp = check_html_get_response(client, url, status_code=301) assert resp["location"] == reverse( "browse-content", query_params={**each_arg, **params} ) def _origin_directory_view_test_helper( client, archive_data, origin_info, origin_visit, snapshot_sizes, origin_branches, origin_releases, root_directory_sha1, directory_entries, visit_id=None, timestamp=None, snapshot_id=None, path=None, ): dirs = [e for e in directory_entries if e["type"] in ("dir", "rev")] files = [e for e in directory_entries if e["type"] == "file"] if not visit_id and not snapshot_id: visit_id = origin_visit["visit"] query_params = {"origin_url": origin_info["url"]} if timestamp: query_params["timestamp"] = timestamp elif visit_id: query_params["visit_id"] = visit_id else: query_params["snapshot"] = snapshot_id if path: query_params["path"] = path url = reverse("browse-origin-directory", query_params=query_params) resp = check_html_get_response( client, url, status_code=200, template_used="browse/directory.html" ) assert_contains(resp, '', count=len(dirs)) assert_contains(resp, '', count=len(files)) if timestamp: query_params["timestamp"] = format_utc_iso_date( parse_iso8601_date_to_utc(timestamp).isoformat(), "%Y-%m-%dT%H:%M:%SZ" ) for d in dirs: if d["type"] == "rev": dir_url = reverse("browse-revision", url_args={"sha1_git": d["target"]}) else: dir_path = d["name"] if path: dir_path = "%s/%s" % (path, d["name"]) query_params["path"] = dir_path dir_url = reverse("browse-origin-directory", query_params=query_params,) assert_contains(resp, dir_url) for f in files: file_path = f["name"] if path: file_path = "%s/%s" % (path, f["name"]) query_params["path"] = file_path file_url = reverse("browse-origin-content", query_params=query_params) assert_contains(resp, file_url) if "path" in query_params: del query_params["path"] root_dir_branch_url = reverse("browse-origin-directory", query_params=query_params) nb_bc_paths = 1 if path: nb_bc_paths = len(path.split("/")) + 1 assert_contains(resp, '
  • ', count=nb_bc_paths) assert_contains( resp, '%s' % (root_dir_branch_url, root_directory_sha1[:7]) ) origin_branches_url = reverse("browse-origin-branches", query_params=query_params) assert_contains(resp, f'href="{escape(origin_branches_url)}"') assert_contains(resp, f"Branches ({snapshot_sizes['revision']})") origin_releases_url = reverse("browse-origin-releases", query_params=query_params) nb_releases = len(origin_releases) if nb_releases > 0: assert_contains(resp, f'href="{escape(origin_releases_url)}"') assert_contains(resp, f"Releases ({snapshot_sizes['release']})") if path: query_params["path"] = path assert_contains(resp, '
  • ', count=len(origin_branches)) for branch in origin_branches: query_params["branch"] = branch["name"] root_dir_branch_url = reverse( "browse-origin-directory", query_params=query_params ) assert_contains(resp, '' % root_dir_branch_url) assert_contains(resp, '
  • ', count=len(origin_releases)) query_params["branch"] = None for release in origin_releases: query_params["release"] = release["name"] root_dir_release_url = reverse( "browse-origin-directory", query_params=query_params ) assert_contains(resp, 'href="%s"' % root_dir_release_url) assert_contains(resp, "vault-cook-directory") assert_contains(resp, "vault-cook-revision") snapshot = archive_data.snapshot_get(origin_visit["snapshot"]) head_rev_id = archive_data.snapshot_get_head(snapshot) swhid_context = { "origin": origin_info["url"], "visit": gen_swhid(ObjectType.SNAPSHOT, snapshot["id"]), "anchor": gen_swhid(ObjectType.REVISION, head_rev_id), "path": f"/{path}" if path else None, } swh_dir_id = gen_swhid( ObjectType.DIRECTORY, directory_entries[0]["dir_id"], metadata=swhid_context ) swh_dir_id_url = reverse("browse-swhid", url_args={"swhid": swh_dir_id}) assert_contains(resp, swh_dir_id) assert_contains(resp, swh_dir_id_url) assert_contains(resp, "swh-take-new-snapshot") _check_origin_link(resp, origin_info["url"]) assert_not_contains(resp, "swh-metadata-popover") def _check_origin_link(resp, origin_url): browse_origin_url = reverse( "browse-origin", query_params={"origin_url": origin_url} ) assert_contains(resp, f'href="{browse_origin_url}"') def test_browse_pull_request_branch( client, archive_data, origin_with_pull_request_branches ): origin_url = origin_with_pull_request_branches.url snapshot = archive_data.snapshot_get_latest(origin_url) pr_branch = random.choice( [ branch for branch in snapshot["branches"].keys() if branch.startswith("refs/pull/") ] ) url = reverse( "browse-origin-directory", query_params={"origin_url": origin_url, "branch": pr_branch}, ) check_html_get_response( client, url, status_code=200, template_used="browse/directory.html" ) diff --git a/swh/web/tests/common/test_archive.py b/swh/web/tests/common/test_archive.py index 4870768a..80b76396 100644 --- a/swh/web/tests/common/test_archive.py +++ b/swh/web/tests/common/test_archive.py @@ -1,1212 +1,1215 @@ # Copyright (C) 2015-2022 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 collections import defaultdict import datetime import hashlib import itertools import random from hypothesis import given, settings import pytest from swh.model.from_disk import DentryPerms from swh.model.hashutil import hash_to_bytes, hash_to_hex from swh.model.model import ( Directory, DirectoryEntry, Origin, OriginVisit, OriginVisitStatus, Revision, Snapshot, SnapshotBranch, TargetType, ) from swh.model.swhids import ObjectType from swh.storage.utils import now from swh.web.common import archive from swh.web.common.exc import BadInputExc, NotFoundExc from swh.web.common.typing import OriginInfo, PagedResult from swh.web.tests.conftest import ctags_json_missing, fossology_missing from swh.web.tests.data import random_content, random_sha1 from swh.web.tests.strategies import new_origin, new_revision, visit_dates def test_lookup_multiple_hashes_all_present(contents): input_data = [] expected_output = [] for cnt in contents: input_data.append({"sha1": cnt["sha1"]}) expected_output.append({"sha1": cnt["sha1"], "found": True}) assert archive.lookup_multiple_hashes(input_data) == expected_output def test_lookup_multiple_hashes_some_missing(contents, unknown_contents): input_contents = list(itertools.chain(contents, unknown_contents)) random.shuffle(input_contents) input_data = [] expected_output = [] for cnt in input_contents: input_data.append({"sha1": cnt["sha1"]}) expected_output.append({"sha1": cnt["sha1"], "found": cnt in contents}) assert archive.lookup_multiple_hashes(input_data) == expected_output def test_lookup_hash_does_not_exist(): unknown_content_ = random_content() actual_lookup = archive.lookup_hash("sha1_git:%s" % unknown_content_["sha1_git"]) assert actual_lookup == {"found": None, "algo": "sha1_git"} def test_lookup_hash_exist(archive_data, content): actual_lookup = archive.lookup_hash("sha1:%s" % content["sha1"]) content_metadata = archive_data.content_get(content["sha1"]) assert {"found": content_metadata, "algo": "sha1"} == actual_lookup def test_search_hash_does_not_exist(): unknown_content_ = random_content() actual_lookup = archive.search_hash("sha1_git:%s" % unknown_content_["sha1_git"]) assert {"found": False} == actual_lookup def test_search_hash_exist(content): actual_lookup = archive.search_hash("sha1:%s" % content["sha1"]) assert {"found": True} == actual_lookup @pytest.mark.skipif( ctags_json_missing, reason="requires ctags with json output support" ) def test_lookup_content_ctags(indexer_data, contents_with_ctags): content_sha1 = random.choice(contents_with_ctags["sha1s"]) indexer_data.content_add_ctags(content_sha1) actual_ctags = list(archive.lookup_content_ctags("sha1:%s" % content_sha1)) expected_data = list(indexer_data.content_get_ctags(content_sha1)) for ctag in expected_data: ctag["id"] = content_sha1 assert actual_ctags == expected_data def test_lookup_content_ctags_no_hash(): unknown_content_ = random_content() actual_ctags = list( archive.lookup_content_ctags("sha1:%s" % unknown_content_["sha1"]) ) assert actual_ctags == [] def test_lookup_content_filetype(indexer_data, content): indexer_data.content_add_mimetype(content["sha1"]) actual_filetype = archive.lookup_content_filetype(content["sha1"]) expected_filetype = indexer_data.content_get_mimetype(content["sha1"]) assert actual_filetype == expected_filetype def test_lookup_expression(indexer_data, contents_with_ctags): per_page = 10 expected_ctags = [] for content_sha1 in contents_with_ctags["sha1s"]: if len(expected_ctags) == per_page: break indexer_data.content_add_ctags(content_sha1) for ctag in indexer_data.content_get_ctags(content_sha1): if len(expected_ctags) == per_page: break if ctag["name"] == contents_with_ctags["symbol_name"]: del ctag["id"] ctag["sha1"] = content_sha1 expected_ctags.append(ctag) actual_ctags = list( archive.lookup_expression( contents_with_ctags["symbol_name"], last_sha1=None, per_page=10 ) ) assert actual_ctags == expected_ctags def test_lookup_expression_no_result(): expected_ctags = [] actual_ctags = list( archive.lookup_expression("barfoo", last_sha1=None, per_page=10) ) assert actual_ctags == expected_ctags @pytest.mark.skipif(fossology_missing, reason="requires fossology-nomossa installed") def test_lookup_content_license(indexer_data, content): indexer_data.content_add_license(content["sha1"]) actual_license = archive.lookup_content_license(content["sha1"]) expected_license = indexer_data.content_get_license(content["sha1"]) assert actual_license == expected_license def test_stat_counters(archive_data): actual_stats = archive.stat_counters() assert actual_stats == archive_data.stat_counters() @given(new_origin(), visit_dates()) def test_lookup_origin_visits(subtest, new_origin, visit_dates): # ensure archive_data fixture will be reset between each hypothesis # example test run @subtest def test_inner(archive_data): archive_data.origin_add([new_origin]) archive_data.origin_visit_add( [ OriginVisit(origin=new_origin.url, date=ts, type="git",) for ts in visit_dates ] ) actual_origin_visits = list( archive.lookup_origin_visits(new_origin.url, per_page=100) ) expected_visits = archive_data.origin_visit_get(new_origin.url) for expected_visit in expected_visits: expected_visit["origin"] = new_origin.url assert actual_origin_visits == expected_visits @given(new_origin(), visit_dates()) def test_lookup_origin_visit(archive_data, new_origin, visit_dates): archive_data.origin_add([new_origin]) visits = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=ts, type="git",) for ts in visit_dates] ) visit = random.choice(visits).visit actual_origin_visit = archive.lookup_origin_visit(new_origin.url, visit) expected_visit = dict(archive_data.origin_visit_get_by(new_origin.url, visit)) assert actual_origin_visit == expected_visit @given(new_origin(), visit_dates()) @settings(max_examples=1) def test_origin_visit_find_by_date_no_result(archive_data, new_origin, visit_dates): """No visit registered in storage for an origin should return no visit""" archive_data.origin_add([new_origin]) for visit_date in visit_dates: # No visit yet, so nothing will get returned actual_origin_visit_status = archive.origin_visit_find_by_date( new_origin.url, visit_date ) assert actual_origin_visit_status is None @settings(max_examples=1) @given(new_origin()) def test_origin_visit_find_by_date(archive_data, new_origin): # Add origin and two visits archive_data.origin_add([new_origin]) pivot_date = now() # First visit one hour before pivot date first_visit_date = pivot_date - datetime.timedelta(hours=1) # Second visit two hours after pivot date second_visit_date = pivot_date + datetime.timedelta(hours=2) visits = archive_data.origin_visit_add( [ OriginVisit(origin=new_origin.url, date=visit_date, type="git",) for visit_date in [first_visit_date, second_visit_date] ] ) # Finalize visits visit_statuses = [] for visit in visits: visit_statuses.append( OriginVisitStatus( origin=new_origin.url, visit=visit.visit, date=visit.date + datetime.timedelta(hours=1), type=visit.type, status="full", snapshot=None, ) ) archive_data.origin_visit_status_add(visit_statuses) # Check correct visit is returned when searching by date - for search_date, expected_visit in [ - (first_visit_date, 1), - (pivot_date, 2), - (second_visit_date, 2), + for search_date, greater_or_equal, expected_visit in [ + (first_visit_date, True, 1), + (pivot_date, True, 2), + (pivot_date, False, 1), + (second_visit_date, True, 2), ]: - origin_visit = archive.origin_visit_find_by_date(new_origin.url, search_date) + origin_visit = archive.origin_visit_find_by_date( + new_origin.url, search_date, greater_or_equal + ) assert origin_visit["visit"] == expected_visit @given(new_origin()) def test_lookup_origin(archive_data, new_origin): archive_data.origin_add([new_origin]) actual_origin = archive.lookup_origin({"url": new_origin.url}) expected_origin = archive_data.origin_get([new_origin.url])[0] assert actual_origin == expected_origin def test_lookup_origin_snapshots(archive_data, origin_with_multiple_visits): origin_url = origin_with_multiple_visits["url"] visits = archive_data.origin_visit_get(origin_url) origin_snapshots = archive.lookup_origin_snapshots(origin_with_multiple_visits) assert set(origin_snapshots) == {v["snapshot"] for v in visits} def test_lookup_release_ko_id_checksum_not_a_sha1(invalid_sha1): with pytest.raises(BadInputExc) as e: archive.lookup_release(invalid_sha1) assert e.match("Invalid checksum") def test_lookup_release_ko_id_checksum_too_long(sha256): with pytest.raises(BadInputExc) as e: archive.lookup_release(sha256) assert e.match("Only sha1_git is supported.") def test_lookup_release_multiple(archive_data, releases): actual_releases = list(archive.lookup_release_multiple(releases)) expected_releases = [] for release_id in releases: release_info = archive_data.release_get(release_id) expected_releases.append(release_info) assert actual_releases == expected_releases def test_lookup_release_multiple_none_found(): unknown_releases_ = [random_sha1(), random_sha1(), random_sha1()] actual_releases = list(archive.lookup_release_multiple(unknown_releases_)) assert actual_releases == [None] * len(unknown_releases_) def test_lookup_directory_with_path_not_found(directory): path = "some/invalid/path/here" with pytest.raises(NotFoundExc) as e: archive.lookup_directory_with_path(directory, path) assert e.match( f"Directory entry with path {path} from root directory {directory} not found" ) def test_lookup_directory_with_path_found(archive_data, directory): directory_content = archive_data.directory_ls(directory) directory_entry = random.choice(directory_content) path = directory_entry["name"] actual_result = archive.lookup_directory_with_path(directory, path) assert actual_result == directory_entry def test_lookup_release(archive_data, release): actual_release = archive.lookup_release(release) assert actual_release == archive_data.release_get(release) def test_lookup_revision_with_context_ko_not_a_sha1(revision, invalid_sha1, sha256): sha1_git_root = revision sha1_git = invalid_sha1 with pytest.raises(BadInputExc) as e: archive.lookup_revision_with_context(sha1_git_root, sha1_git) assert e.match("Invalid checksum query string") sha1_git = sha256 with pytest.raises(BadInputExc) as e: archive.lookup_revision_with_context(sha1_git_root, sha1_git) assert e.match("Only sha1_git is supported") def test_lookup_revision_with_context_ko_sha1_git_does_not_exist( revision, unknown_revision ): sha1_git_root = revision sha1_git = unknown_revision with pytest.raises(NotFoundExc) as e: archive.lookup_revision_with_context(sha1_git_root, sha1_git) assert e.match("Revision %s not found" % sha1_git) def test_lookup_revision_with_context_ko_root_sha1_git_does_not_exist( revision, unknown_revision ): sha1_git_root = unknown_revision sha1_git = revision with pytest.raises(NotFoundExc) as e: archive.lookup_revision_with_context(sha1_git_root, sha1_git) assert e.match("Revision root %s not found" % sha1_git_root) def test_lookup_revision_with_context(archive_data, ancestor_revisions): sha1_git = ancestor_revisions["sha1_git"] root_sha1_git = ancestor_revisions["sha1_git_root"] for sha1_git_root in (root_sha1_git, {"id": hash_to_bytes(root_sha1_git)}): actual_revision = archive.lookup_revision_with_context(sha1_git_root, sha1_git) children = [] for rev in archive_data.revision_log(root_sha1_git): for p_rev in rev["parents"]: p_rev_hex = hash_to_hex(p_rev) if p_rev_hex == sha1_git: children.append(rev["id"]) expected_revision = archive_data.revision_get(sha1_git) expected_revision["children"] = children assert actual_revision == expected_revision def test_lookup_revision_with_context_ko(non_ancestor_revisions): sha1_git = non_ancestor_revisions["sha1_git"] root_sha1_git = non_ancestor_revisions["sha1_git_root"] with pytest.raises(NotFoundExc) as e: archive.lookup_revision_with_context(root_sha1_git, sha1_git) assert e.match("Revision %s is not an ancestor of %s" % (sha1_git, root_sha1_git)) def test_lookup_directory_with_revision_not_found(): unknown_revision_ = random_sha1() with pytest.raises(NotFoundExc) as e: archive.lookup_directory_with_revision(unknown_revision_) assert e.match("Revision %s not found" % unknown_revision_) @given(new_revision()) def test_lookup_directory_with_revision_unknown_content(archive_data, new_revision): unknown_content_ = random_content() dir_path = "README.md" # A directory that points to unknown content dir = Directory( entries=( DirectoryEntry( name=bytes(dir_path.encode("utf-8")), type="file", target=hash_to_bytes(unknown_content_["sha1_git"]), perms=DentryPerms.content, ), ) ) # Create a revision that points to a directory # Which points to unknown content new_revision = new_revision.to_dict() new_revision["directory"] = dir.id del new_revision["id"] new_revision = Revision.from_dict(new_revision) # Add the directory and revision in mem archive_data.directory_add([dir]) archive_data.revision_add([new_revision]) new_revision_id = hash_to_hex(new_revision.id) with pytest.raises(NotFoundExc) as e: archive.lookup_directory_with_revision(new_revision_id, dir_path) assert e.match("Content not found for revision %s" % new_revision_id) def test_lookup_directory_with_revision_ko_path_to_nowhere(revision): invalid_path = "path/to/something/unknown" with pytest.raises(NotFoundExc) as e: archive.lookup_directory_with_revision(revision, invalid_path) assert e.match("Directory or File") assert e.match(invalid_path) assert e.match("revision %s" % revision) assert e.match("not found") def test_lookup_directory_with_revision_submodules( archive_data, revision_with_submodules ): rev_sha1_git = revision_with_submodules["rev_sha1_git"] rev_dir_path = revision_with_submodules["rev_dir_rev_path"] actual_data = archive.lookup_directory_with_revision(rev_sha1_git, rev_dir_path) revision = archive_data.revision_get(revision_with_submodules["rev_sha1_git"]) directory = archive_data.directory_ls(revision["directory"]) rev_entry = next(e for e in directory if e["name"] == rev_dir_path) expected_data = { "content": archive_data.revision_get(rev_entry["target"]), "path": rev_dir_path, "revision": rev_sha1_git, "type": "rev", } assert actual_data == expected_data def test_lookup_directory_with_revision_without_path(archive_data, revision): actual_directory_entries = archive.lookup_directory_with_revision(revision) revision_data = archive_data.revision_get(revision) expected_directory_entries = archive_data.directory_ls(revision_data["directory"]) assert actual_directory_entries["type"] == "dir" assert actual_directory_entries["content"] == expected_directory_entries def test_lookup_directory_with_revision_with_path(archive_data, revision): rev_data = archive_data.revision_get(revision) dir_entries = [ e for e in archive_data.directory_ls(rev_data["directory"]) if e["type"] in ("file", "dir") ] expected_dir_entry = random.choice(dir_entries) actual_dir_entry = archive.lookup_directory_with_revision( revision, expected_dir_entry["name"] ) assert actual_dir_entry["type"] == expected_dir_entry["type"] assert actual_dir_entry["revision"] == revision assert actual_dir_entry["path"] == expected_dir_entry["name"] if actual_dir_entry["type"] == "file": del actual_dir_entry["content"]["checksums"]["blake2s256"] for key in ("checksums", "status", "length"): assert actual_dir_entry["content"][key] == expected_dir_entry[key] else: sub_dir_entries = archive_data.directory_ls(expected_dir_entry["target"]) assert actual_dir_entry["content"] == sub_dir_entries def test_lookup_directory_with_revision_with_path_to_file_and_data( archive_data, revision ): rev_data = archive_data.revision_get(revision) dir_entries = [ e for e in archive_data.directory_ls(rev_data["directory"]) if e["type"] == "file" ] expected_dir_entry = random.choice(dir_entries) expected_data = archive_data.content_get_data( expected_dir_entry["checksums"]["sha1"] ) actual_dir_entry = archive.lookup_directory_with_revision( revision, expected_dir_entry["name"], with_data=True ) assert actual_dir_entry["type"] == expected_dir_entry["type"] assert actual_dir_entry["revision"] == revision assert actual_dir_entry["path"] == expected_dir_entry["name"] del actual_dir_entry["content"]["checksums"]["blake2s256"] for key in ("checksums", "status", "length"): assert actual_dir_entry["content"][key] == expected_dir_entry[key] assert actual_dir_entry["content"]["data"] == expected_data["data"] def test_lookup_revision(archive_data, revision): actual_revision = archive.lookup_revision(revision) assert actual_revision == archive_data.revision_get(revision) @given(new_revision()) def test_lookup_revision_invalid_msg(archive_data, new_revision): new_revision = new_revision.to_dict() new_revision["message"] = b"elegant fix for bug \xff" archive_data.revision_add([Revision.from_dict(new_revision)]) revision = archive.lookup_revision(hash_to_hex(new_revision["id"])) assert revision["message"] == "elegant fix for bug \\xff" assert "message" in revision["decoding_failures"] @given(new_revision()) def test_lookup_revision_msg_ok(archive_data, new_revision): archive_data.revision_add([new_revision]) revision_message = archive.lookup_revision_message(hash_to_hex(new_revision.id)) assert revision_message == {"message": new_revision.message} def test_lookup_revision_msg_no_rev(): unknown_revision_ = random_sha1() with pytest.raises(NotFoundExc) as e: archive.lookup_revision_message(unknown_revision_) assert e.match("Revision with sha1_git %s not found." % unknown_revision_) def test_lookup_revision_multiple(archive_data, revisions): actual_revisions = list(archive.lookup_revision_multiple(revisions)) expected_revisions = [] for rev in revisions: expected_revisions.append(archive_data.revision_get(rev)) assert actual_revisions == expected_revisions def test_lookup_revision_multiple_none_found(): unknown_revisions_ = [random_sha1(), random_sha1(), random_sha1()] actual_revisions = list(archive.lookup_revision_multiple(unknown_revisions_)) assert actual_revisions == [None] * len(unknown_revisions_) def test_lookup_revision_log(archive_data, revision): actual_revision_log = list(archive.lookup_revision_log(revision, limit=25)) expected_revision_log = archive_data.revision_log(revision, limit=25) assert actual_revision_log == expected_revision_log def _get_origin_branches(archive_data, origin): origin_visit = archive_data.origin_visit_get(origin["url"])[-1] snapshot = archive_data.snapshot_get(origin_visit["snapshot"]) branches = { k: v for (k, v) in snapshot["branches"].items() if v["target_type"] == "revision" } return branches def test_lookup_revision_log_by(archive_data, origin): branches = _get_origin_branches(archive_data, origin) branch_name = random.choice(list(branches.keys())) actual_log = list( archive.lookup_revision_log_by(origin["url"], branch_name, None, limit=25) ) expected_log = archive_data.revision_log(branches[branch_name]["target"], limit=25) assert actual_log == expected_log def test_lookup_revision_log_by_notfound(origin): with pytest.raises(NotFoundExc): archive.lookup_revision_log_by( origin["url"], "unknown_branch_name", None, limit=100 ) def test_lookup_content_raw_not_found(): unknown_content_ = random_content() with pytest.raises(NotFoundExc) as e: archive.lookup_content_raw("sha1:" + unknown_content_["sha1"]) assert e.match( "Content with %s checksum equals to %s not found!" % ("sha1", unknown_content_["sha1"]) ) def test_lookup_content_raw(archive_data, content): actual_content = archive.lookup_content_raw("sha256:%s" % content["sha256"]) expected_content = archive_data.content_get_data(content["sha1"]) assert actual_content == expected_content def test_lookup_empty_content_raw(empty_content): content_raw = archive.lookup_content_raw(f"sha1_git:{empty_content['sha1_git']}") assert content_raw["data"] == b"" def test_lookup_content_not_found(): unknown_content_ = random_content() with pytest.raises(NotFoundExc) as e: archive.lookup_content("sha1:%s" % unknown_content_["sha1"]) assert e.match( "Content with %s checksum equals to %s not found!" % ("sha1", unknown_content_["sha1"]) ) def test_lookup_content_with_sha1(archive_data, content): actual_content = archive.lookup_content(f"sha1:{content['sha1']}") expected_content = archive_data.content_get(content["sha1"]) assert actual_content == expected_content def test_lookup_content_with_sha256(archive_data, content): actual_content = archive.lookup_content(f"sha256:{content['sha256']}") expected_content = archive_data.content_get(content["sha1"]) assert actual_content == expected_content def test_lookup_directory_bad_checksum(): with pytest.raises(BadInputExc): archive.lookup_directory("directory_id") def test_lookup_directory_not_found(): unknown_directory_ = random_sha1() with pytest.raises(NotFoundExc) as e: archive.lookup_directory(unknown_directory_) assert e.match("Directory with sha1_git %s not found" % unknown_directory_) def test_lookup_directory(archive_data, directory): actual_directory_ls = list(archive.lookup_directory(directory)) expected_directory_ls = archive_data.directory_ls(directory) assert actual_directory_ls == expected_directory_ls def test_lookup_directory_empty(empty_directory): actual_directory_ls = list(archive.lookup_directory(empty_directory)) assert actual_directory_ls == [] def test_lookup_revision_by_nothing_found(origin): with pytest.raises(NotFoundExc): archive.lookup_revision_by(origin["url"], "invalid-branch-name") def test_lookup_revision_by(archive_data, origin): branches = _get_origin_branches(archive_data, origin) branch_name = random.choice(list(branches.keys())) actual_revision = archive.lookup_revision_by(origin["url"], branch_name) expected_revision = archive_data.revision_get(branches[branch_name]["target"]) assert actual_revision == expected_revision def test_lookup_revision_with_context_by_ko(origin, revision): with pytest.raises(NotFoundExc): archive.lookup_revision_with_context_by( origin["url"], "invalid-branch-name", None, revision ) def test_lookup_revision_with_context_by(archive_data, origin): branches = _get_origin_branches(archive_data, origin) branch_name = random.choice(list(branches.keys())) root_rev = branches[branch_name]["target"] root_rev_log = archive_data.revision_log(root_rev) children = defaultdict(list) for rev in root_rev_log: for rev_p in rev["parents"]: children[rev_p].append(rev["id"]) rev = root_rev_log[-1]["id"] actual_root_rev, actual_rev = archive.lookup_revision_with_context_by( origin["url"], branch_name, None, rev ) expected_root_rev = archive_data.revision_get(root_rev) expected_rev = archive_data.revision_get(rev) expected_rev["children"] = children[rev] assert actual_root_rev == expected_root_rev assert actual_rev == expected_rev def test_lookup_revision_through_ko_not_implemented(): with pytest.raises(NotImplementedError): archive.lookup_revision_through({"something-unknown": 10}) def test_lookup_revision_through_with_context_by(archive_data, origin): branches = _get_origin_branches(archive_data, origin) branch_name = random.choice(list(branches.keys())) root_rev = branches[branch_name]["target"] root_rev_log = archive_data.revision_log(root_rev) rev = root_rev_log[-1]["id"] assert archive.lookup_revision_through( { "origin_url": origin["url"], "branch_name": branch_name, "ts": None, "sha1_git": rev, } ) == archive.lookup_revision_with_context_by(origin["url"], branch_name, None, rev) def test_lookup_revision_through_with_revision_by(archive_data, origin): branches = _get_origin_branches(archive_data, origin) branch_name = random.choice(list(branches.keys())) assert archive.lookup_revision_through( {"origin_url": origin["url"], "branch_name": branch_name, "ts": None,} ) == archive.lookup_revision_by(origin["url"], branch_name, None) def test_lookup_revision_through_with_context(ancestor_revisions): sha1_git = ancestor_revisions["sha1_git"] sha1_git_root = ancestor_revisions["sha1_git_root"] assert archive.lookup_revision_through( {"sha1_git_root": sha1_git_root, "sha1_git": sha1_git,} ) == archive.lookup_revision_with_context(sha1_git_root, sha1_git) def test_lookup_revision_through_with_revision(revision): assert archive.lookup_revision_through( {"sha1_git": revision} ) == archive.lookup_revision(revision) def test_lookup_directory_through_revision_ko_not_found(revision): with pytest.raises(NotFoundExc): archive.lookup_directory_through_revision( {"sha1_git": revision}, "some/invalid/path" ) def test_lookup_directory_through_revision_ok(archive_data, revision): rev_data = archive_data.revision_get(revision) dir_entries = [ e for e in archive_data.directory_ls(rev_data["directory"]) if e["type"] == "file" ] dir_entry = random.choice(dir_entries) assert archive.lookup_directory_through_revision( {"sha1_git": revision}, dir_entry["name"] ) == (revision, archive.lookup_directory_with_revision(revision, dir_entry["name"])) def test_lookup_directory_through_revision_ok_with_data(archive_data, revision): rev_data = archive_data.revision_get(revision) dir_entries = [ e for e in archive_data.directory_ls(rev_data["directory"]) if e["type"] == "file" ] dir_entry = random.choice(dir_entries) assert archive.lookup_directory_through_revision( {"sha1_git": revision}, dir_entry["name"], with_data=True ) == ( revision, archive.lookup_directory_with_revision( revision, dir_entry["name"], with_data=True ), ) def test_lookup_known_objects( archive_data, content, directory, release, revision, snapshot ): expected = archive_data.content_find(content) assert archive.lookup_object(ObjectType.CONTENT, content["sha1_git"]) == expected expected = archive_data.directory_get(directory) assert archive.lookup_object(ObjectType.DIRECTORY, directory) == expected expected = archive_data.release_get(release) assert archive.lookup_object(ObjectType.RELEASE, release) == expected expected = archive_data.revision_get(revision) assert archive.lookup_object(ObjectType.REVISION, revision) == expected expected = {**archive_data.snapshot_get(snapshot), "next_branch": None} assert archive.lookup_object(ObjectType.SNAPSHOT, snapshot) == expected def test_lookup_unknown_objects( unknown_content, unknown_directory, unknown_release, unknown_revision, unknown_snapshot, ): with pytest.raises(NotFoundExc) as e: archive.lookup_object(ObjectType.CONTENT, unknown_content["sha1_git"]) assert e.match(r"Content.*not found") with pytest.raises(NotFoundExc) as e: archive.lookup_object(ObjectType.DIRECTORY, unknown_directory) assert e.match(r"Directory.*not found") with pytest.raises(NotFoundExc) as e: archive.lookup_object(ObjectType.RELEASE, unknown_release) assert e.match(r"Release.*not found") with pytest.raises(NotFoundExc) as e: archive.lookup_object(ObjectType.REVISION, unknown_revision) assert e.match(r"Revision.*not found") with pytest.raises(NotFoundExc) as e: archive.lookup_object(ObjectType.SNAPSHOT, unknown_snapshot) assert e.match(r"Snapshot.*not found") def test_lookup_invalid_objects(invalid_sha1): with pytest.raises(BadInputExc) as e: archive.lookup_object(ObjectType.CONTENT, invalid_sha1) assert e.match("Invalid hash") with pytest.raises(BadInputExc) as e: archive.lookup_object(ObjectType.DIRECTORY, invalid_sha1) assert e.match("Invalid checksum") with pytest.raises(BadInputExc) as e: archive.lookup_object(ObjectType.RELEASE, invalid_sha1) assert e.match("Invalid checksum") with pytest.raises(BadInputExc) as e: archive.lookup_object(ObjectType.REVISION, invalid_sha1) assert e.match("Invalid checksum") with pytest.raises(BadInputExc) as e: archive.lookup_object(ObjectType.SNAPSHOT, invalid_sha1) assert e.match("Invalid checksum") def test_lookup_missing_hashes_non_present(): missing_cnt = random_sha1() missing_dir = random_sha1() missing_rev = random_sha1() missing_rel = random_sha1() missing_snp = random_sha1() grouped_swhids = { ObjectType.CONTENT: [hash_to_bytes(missing_cnt)], ObjectType.DIRECTORY: [hash_to_bytes(missing_dir)], ObjectType.REVISION: [hash_to_bytes(missing_rev)], ObjectType.RELEASE: [hash_to_bytes(missing_rel)], ObjectType.SNAPSHOT: [hash_to_bytes(missing_snp)], } actual_result = archive.lookup_missing_hashes(grouped_swhids) assert actual_result == { missing_cnt, missing_dir, missing_rev, missing_rel, missing_snp, } def test_lookup_missing_hashes_some_present(content, directory): missing_rev = random_sha1() missing_rel = random_sha1() missing_snp = random_sha1() grouped_swhids = { ObjectType.CONTENT: [hash_to_bytes(content["sha1_git"])], ObjectType.DIRECTORY: [hash_to_bytes(directory)], ObjectType.REVISION: [hash_to_bytes(missing_rev)], ObjectType.RELEASE: [hash_to_bytes(missing_rel)], ObjectType.SNAPSHOT: [hash_to_bytes(missing_snp)], } actual_result = archive.lookup_missing_hashes(grouped_swhids) assert actual_result == {missing_rev, missing_rel, missing_snp} def test_lookup_origin_extra_trailing_slash(origin): origin_info = archive.lookup_origin({"url": f"{origin['url']}/"}) assert origin_info["url"] == origin["url"] def test_lookup_origin_missing_trailing_slash(archive_data): deb_origin = Origin(url="http://snapshot.debian.org/package/r-base/") archive_data.origin_add([deb_origin]) origin_info = archive.lookup_origin({"url": deb_origin.url[:-1]}) assert origin_info["url"] == deb_origin.url def test_lookup_origin_single_slash_after_protocol(archive_data): origin_url = "http://snapshot.debian.org/package/r-base/" malformed_origin_url = "http:/snapshot.debian.org/package/r-base/" archive_data.origin_add([Origin(url=origin_url)]) origin_info = archive.lookup_origin({"url": malformed_origin_url}) assert origin_info["url"] == origin_url @given(new_origin()) def test_lookup_origins_get_by_sha1s(origin, unknown_origin): hasher = hashlib.sha1() hasher.update(origin["url"].encode("utf-8")) origin_info = OriginInfo(url=origin["url"]) origin_sha1 = hasher.hexdigest() hasher = hashlib.sha1() hasher.update(unknown_origin.url.encode("utf-8")) unknown_origin_sha1 = hasher.hexdigest() origins = list(archive.lookup_origins_by_sha1s([origin_sha1])) assert origins == [origin_info] origins = list(archive.lookup_origins_by_sha1s([origin_sha1, origin_sha1])) assert origins == [origin_info, origin_info] origins = list(archive.lookup_origins_by_sha1s([origin_sha1, unknown_origin_sha1])) assert origins == [origin_info, None] def test_search_origin(origin): results = archive.search_origin(url_pattern=origin["url"])[0] assert results == [{"url": origin["url"]}] def test_search_origin_use_ql(mocker, origin): ORIGIN = [{"url": origin["url"]}] mock_archive_search = mocker.patch("swh.web.common.archive.search") mock_archive_search.origin_search.return_value = PagedResult( results=ORIGIN, next_page_token=None, ) query = f"origin = '{origin['url']}'" results = archive.search_origin(url_pattern=query, use_ql=True)[0] assert results == ORIGIN mock_archive_search.origin_search.assert_called_with( query=query, page_token=None, with_visit=False, visit_types=None, limit=50 ) def test_lookup_snapshot_sizes(archive_data, snapshot): branches = archive_data.snapshot_get(snapshot)["branches"] expected_sizes = { "alias": 0, "release": 0, "revision": 0, } for branch_name, branch_info in branches.items(): if branch_info is not None: expected_sizes[branch_info["target_type"]] += 1 assert archive.lookup_snapshot_sizes(snapshot) == expected_sizes def test_lookup_snapshot_sizes_with_filtering(archive_data, revision): rev_id = hash_to_bytes(revision) snapshot = Snapshot( branches={ b"refs/heads/master": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), b"refs/heads/incoming": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), b"refs/pull/1": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), b"refs/pull/2": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), }, ) archive_data.snapshot_add([snapshot]) expected_sizes = {"alias": 0, "release": 0, "revision": 2} assert ( archive.lookup_snapshot_sizes( snapshot.id.hex(), branch_name_exclude_prefix="refs/pull/" ) == expected_sizes ) def test_lookup_snapshot_alias(snapshot): resolved_alias = archive.lookup_snapshot_alias(snapshot, "HEAD") assert resolved_alias is not None assert resolved_alias["target_type"] == "revision" assert resolved_alias["target"] is not None def test_lookup_snapshot_missing(revision): with pytest.raises(NotFoundExc): archive.lookup_snapshot(revision) def test_lookup_snapshot_empty_branch_list(archive_data, revision): rev_id = hash_to_bytes(revision) snapshot = Snapshot( branches={ b"refs/heads/master": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), }, ) archive_data.snapshot_add([snapshot]) # FIXME; This test will change once the inconsistency in storage is fixed # postgres backend returns None in case of a missing branch whereas the # in-memory implementation (used in tests) returns a data structure; # hence the inconsistency branches = archive.lookup_snapshot( hash_to_hex(snapshot.id), branch_name_include_substring="non-existing", )["branches"] assert not branches def test_lookup_snapshot_branch_names_filtering(archive_data, revision): rev_id = hash_to_bytes(revision) snapshot = Snapshot( branches={ b"refs/heads/master": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), b"refs/heads/incoming": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), b"refs/pull/1": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), b"refs/pull/2": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), "non_ascii_name_é".encode(): SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), }, ) archive_data.snapshot_add([snapshot]) for include_pattern, exclude_prefix, nb_results in ( ("pull", None, 2), ("incoming", None, 1), ("é", None, 1), (None, "refs/heads/", 3), ("refs", "refs/heads/master", 3), ): branches = archive.lookup_snapshot( hash_to_hex(snapshot.id), branch_name_include_substring=include_pattern, branch_name_exclude_prefix=exclude_prefix, )["branches"] assert len(branches) == nb_results for branch_name in branches: if include_pattern: assert include_pattern in branch_name if exclude_prefix: assert not branch_name.startswith(exclude_prefix) def test_lookup_snapshot_branch_names_filtering_paginated( archive_data, directory, revision ): pattern = "foo" nb_branches_by_target_type = 10 branches = {} for i in range(nb_branches_by_target_type): branches[f"branch/directory/bar{i}".encode()] = SnapshotBranch( target=hash_to_bytes(directory), target_type=TargetType.DIRECTORY, ) branches[f"branch/revision/bar{i}".encode()] = SnapshotBranch( target=hash_to_bytes(revision), target_type=TargetType.REVISION, ) branches[f"branch/directory/{pattern}{i}".encode()] = SnapshotBranch( target=hash_to_bytes(directory), target_type=TargetType.DIRECTORY, ) branches[f"branch/revision/{pattern}{i}".encode()] = SnapshotBranch( target=hash_to_bytes(revision), target_type=TargetType.REVISION, ) snapshot = Snapshot(branches=branches) archive_data.snapshot_add([snapshot]) branches_count = nb_branches_by_target_type // 2 for target_type in ( ObjectType.DIRECTORY.name.lower(), ObjectType.REVISION.name.lower(), ): partial_branches = archive.lookup_snapshot( hash_to_hex(snapshot.id), target_types=[target_type], branches_count=branches_count, branch_name_include_substring=pattern, ) branches = partial_branches["branches"] assert len(branches) == branches_count for branch_name, branch_data in branches.items(): assert pattern in branch_name assert branch_data["target_type"] == target_type for i in range(branches_count): assert f"branch/{target_type}/{pattern}{i}" in branches assert ( partial_branches["next_branch"] == f"branch/{target_type}/{pattern}{branches_count}" ) partial_branches = archive.lookup_snapshot( hash_to_hex(snapshot.id), target_types=[target_type], branches_from=partial_branches["next_branch"], branch_name_include_substring=pattern, ) branches = partial_branches["branches"] assert len(branches) == branches_count for branch_name, branch_data in branches.items(): assert pattern in branch_name assert branch_data["target_type"] == target_type for i in range(branches_count, 2 * branches_count): assert f"branch/{target_type}/{pattern}{i}" in branches assert partial_branches["next_branch"] is None diff --git a/swh/web/tests/common/test_origin_visits.py b/swh/web/tests/common/test_origin_visits.py index d3a54b08..f6435b6f 100644 --- a/swh/web/tests/common/test_origin_visits.py +++ b/swh/web/tests/common/test_origin_visits.py @@ -1,226 +1,226 @@ -# Copyright (C) 2018-2021 The Software Heritage developers +# Copyright (C) 2018-2022 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 datetime import timedelta from hypothesis import given, settings import iso8601 import pytest from swh.model.model import OriginVisit, OriginVisitStatus from swh.storage.utils import now from swh.web.common.exc import NotFoundExc from swh.web.common.origin_visits import get_origin_visit, get_origin_visits from swh.web.common.typing import OriginInfo from swh.web.tests.strategies import new_origin, new_snapshots @settings(max_examples=1) @given(new_origin(), new_snapshots(3)) def test_get_origin_visits(mocker, archive_data, new_origin, new_snapshots): from swh.web.common import archive mocker.patch.object(archive, "MAX_LIMIT", 2) archive_data.origin_add([new_origin]) archive_data.snapshot_add(new_snapshots) for i, snapshot in enumerate(new_snapshots): visit_date = now() + timedelta(days=i * 10) visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=visit_date, type="git",)] )[0] visit_status = OriginVisitStatus( origin=new_origin.url, visit=visit.visit, date=visit_date + timedelta(minutes=5), status="full", snapshot=snapshot.id, ) archive_data.origin_visit_status_add([visit_status]) origin_visits = get_origin_visits(new_origin.to_dict()) assert len(origin_visits) == len(new_snapshots) @given(new_origin(), new_snapshots(5)) def test_get_origin_visit(archive_data, new_origin, new_snapshots): archive_data.origin_add([new_origin]) archive_data.snapshot_add(new_snapshots) visits = [] for i, visit_date in enumerate( map( iso8601.parse_date, [ "2015-07-09T21:09:24+00:00", "2016-02-23T18:05:23.312045+00:00", "2016-03-28T01:35:06.554111+00:00", "2016-06-18T01:22:24.808485+00:00", "2016-08-14T12:10:00.536702+00:00", ], ) ): visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=visit_date, type="git",)] )[0] visits.append(visit) visit_status = OriginVisitStatus( origin=new_origin.url, visit=visit.visit, date=visit_date + timedelta(minutes=5), status="full", snapshot=new_snapshots[i].id, ) archive_data.origin_visit_status_add([visit_status]) origin_info = new_origin.to_dict() visit_id = visits[-1].visit + 1 with pytest.raises(NotFoundExc) as e: visit = get_origin_visit(origin_info, visit_id=visit_id) - assert e.match("Visit with id %s" % visit_id) - assert e.match("url %s" % origin_info["url"]) + assert e.match("visit with id %s" % visit_id) + assert e.match("Origin %s" % origin_info["url"]) visit_id = visits[1].visit visit = get_origin_visit(origin_info, visit_id=visit_id) assert visit == archive_data.origin_visit_get_by(new_origin.url, visit_id=visit_id) visit = get_origin_visit(origin_info, visit_ts="2016-02-23T18:05:23.312045+00:00") assert visit == archive_data.origin_visit_get_by(new_origin.url, visit_id=visit_id) visit = get_origin_visit(origin_info, visit_ts="2016-02-20") assert visit == archive_data.origin_visit_get_by(new_origin.url, visit_id=visit_id) visit_id = visits[3].visit visit = get_origin_visit(origin_info, visit_ts="2016-06-18T01:22") assert visit == archive_data.origin_visit_get_by(new_origin.url, visit_id=visit_id) visit = get_origin_visit(origin_info, visit_ts="2016-06-18 01:22") assert visit == archive_data.origin_visit_get_by(new_origin.url, visit_id=visit_id) visit_id = visits[0].visit visit = get_origin_visit(origin_info, visit_ts="2014-01-01") assert visit == archive_data.origin_visit_get_by(new_origin.url, visit_id=visit_id) visit_id = visits[-1].visit visit = get_origin_visit(origin_info, visit_ts="2018-01-01") assert visit == archive_data.origin_visit_get_by(new_origin.url, visit_id=visit_id) @given(new_origin(), new_snapshots(6)) def test_get_origin_visit_return_first_valid_full_visit( archive_data, new_origin, new_snapshots ): visits = [] archive_data.origin_add([new_origin]) # create 6 visits, the first three have full status while the # last three have partial status and set a null snapshot for # the last four visits for i, snp in enumerate(new_snapshots): visit_date = now() + timedelta(days=i * 10) visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=visit_date, type="git",)] )[0] archive_data.snapshot_add([new_snapshots[i]]) visit_status = OriginVisitStatus( origin=new_origin.url, visit=visit.visit, date=visit_date + timedelta(minutes=5), status="full" if i < 3 else "partial", snapshot=new_snapshots[i].id if i < 2 else None, ) if i < 2: archive_data.origin_visit_status_add([visit_status]) visits.append(visit.visit) # should return the second visit expected_visit = archive_data.origin_visit_get_by(new_origin.url, visits[1]) assert get_origin_visit((OriginInfo(url=new_origin.url))) == expected_visit @given(new_origin(), new_snapshots(6)) def test_get_origin_visit_non_resolvable_snapshots( archive_data, new_origin, new_snapshots ): visits = [] archive_data.origin_add([new_origin]) # create 6 full visits, the first three have resolvable snapshots # while the last three have non resolvable snapshots for i, snp in enumerate(new_snapshots): visit_date = now() + timedelta(days=i * 10) visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=visit_date, type="git",)] )[0] archive_data.snapshot_add([new_snapshots[i]]) visit_status = OriginVisitStatus( origin=new_origin.url, visit=visit.visit, date=visit_date + timedelta(minutes=5), status="full", snapshot=new_snapshots[i].id, ) if i < 3: archive_data.origin_visit_status_add([visit_status]) visits.append(visit.visit) # should return the third visit expected_visit = archive_data.origin_visit_get_by(new_origin.url, visits[2]) assert get_origin_visit((OriginInfo(url=new_origin.url))) == expected_visit @given(new_origin(), new_snapshots(6)) def test_get_origin_visit_return_first_valid_partial_visit( archive_data, new_origin, new_snapshots ): visits = [] archive_data.origin_add([new_origin]) # create 6 visits, the first three have full status but null snapshot # while the last three have partial status with valid snapshot for i, snp in enumerate(new_snapshots): visit_date = now() + timedelta(days=i * 10) visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=visit_date, type="git",)] )[0] archive_data.snapshot_add([new_snapshots[i]]) visit_status = OriginVisitStatus( origin=new_origin.url, visit=visit.visit, date=visit_date + timedelta(minutes=5), status="full" if i < 3 else "partial", snapshot=new_snapshots[i].id if i > 2 else None, ) if i > 2: archive_data.origin_visit_status_add([visit_status]) visits.append(visit.visit) # should return the last visit expected_visit = archive_data.origin_visit_get_by(new_origin.url, visits[-1]) assert get_origin_visit((OriginInfo(url=new_origin.url))) == expected_visit def test_get_origin_visit_latest_snapshot(mocker, origin_with_multiple_visits): origin_visits = get_origin_visits(origin_with_multiple_visits) first_visit = origin_visits[0] latest_visit = origin_visits[-1] mock_get_origin_visits = mocker.patch( "swh.web.common.origin_visits.get_origin_visits" ) mock_get_origin_visits.return_value = origin_visits visit = get_origin_visit( origin_with_multiple_visits, snapshot_id=latest_visit["snapshot"] ) assert visit == latest_visit assert not mock_get_origin_visits.called visit = get_origin_visit( origin_with_multiple_visits, snapshot_id=first_visit["snapshot"] ) assert visit == first_visit assert mock_get_origin_visits.called