diff --git a/swh/web/browse/views/revision.py b/swh/web/browse/views/revision.py index 6177f52a..ae4714f4 100644 --- a/swh/web/browse/views/revision.py +++ b/swh/web/browse/views/revision.py @@ -1,604 +1,607 @@ # Copyright (C) 2017-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import hashlib import json import textwrap from django.http import HttpResponse from django.shortcuts import render from django.template.defaultfilters import filesizeformat from django.utils.safestring import mark_safe from swh.model.identifiers import ( persistent_identifier, CONTENT, DIRECTORY, REVISION, SNAPSHOT, ) 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_link, gen_revision_link, gen_revision_url, get_revision_log_url, get_directory_entries, gen_directory_link, request_content, prepare_content_for_display, content_display_max_size, gen_snapshot_link, get_readme_to_display, format_log_entries, gen_person_mail_link, ) from swh.web.common import service from swh.web.common.exc import NotFoundExc, handle_view_exception from swh.web.common.identifiers import get_swhids_info from swh.web.common.typing import RevisionMetadata, SWHObjectInfo from swh.web.common.utils import ( reverse, format_utc_iso_date, gen_path_info, 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 panel_diff_link = "#panel_" + diff_id if change["type"] == "modify": change["content_url"] = _gen_content_url( revision, to_query_string, change["to_path"], snapshot_context ) changes_msg.append( "modified: %s" % _gen_diff_link(i, panel_diff_link, change["to_path"]) ) elif change["type"] == "insert": change["content_url"] = _gen_content_url( revision, to_query_string, change["to_path"], snapshot_context ) changes_msg.append( "new file: %s" % _gen_diff_link(i, panel_diff_link, change["to_path"]) ) elif change["type"] == "delete": parent = service.lookup_revision(revision["parents"][0]) change["content_url"] = _gen_content_url( parent, from_query_string, change["from_path"], snapshot_context ) changes_msg.append( "deleted: %s" % _gen_diff_link(i, panel_diff_link, change["from_path"]) ) elif change["type"] == "rename": change["content_url"] = _gen_content_url( revision, to_query_string, change["to_path"], snapshot_context ) link_text = change["from_path"] + " → " + change["to_path"] changes_msg.append( "renamed: %s" % _gen_diff_link(i, panel_diff_link, link_text) ) if not changes: changes_msg.append("No changes") return mark_safe("\n".join(changes_msg)) @browse_route( r"revision/(?P[0-9a-f]+)/diff/", view_name="diff-revision", checksum_args=["sha1_git"], ) def _revision_diff(request, sha1_git): """ Browse internal endpoint to compute revision diff """ try: revision = service.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 ) except Exception as exc: return handle_view_exception(request, exc) changes = service.diff_revision(sha1_git) changes_msg = _gen_revision_changes_list(revision, changes, snapshot_context) diff_data = { "total_nb_changes": len(changes), "changes": changes[:_max_displayed_file_diffs], "changes_msg": changes_msg, } diff_data_json = json.dumps(diff_data, separators=(",", ": ")) return HttpResponse(diff_data_json, content_type="application/json") NB_LOG_ENTRIES = 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/` """ try: origin_url = request.GET.get("origin_url") snapshot_id = request.GET.get("snapshot") snapshot_context = None if origin_url or snapshot_id: 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"), 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 = service.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 = service.lookup_revision_multiple(revs) request.session[session_key] = { "rev_log": rev_log, "revs_walker_state": revs_walker_state, } except Exception as exc: return handle_view_exception(request, exc) 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, }, ) 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, }, ) revision_log_data = format_log_entries(revision_log, per_page) swh_rev_id = persistent_identifier("revision", 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_menu": 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)/`. """ try: revision = service.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") 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 = None, None 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, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=sha1_git, ) 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) if path: file_info = service.lookup_directory_with_path(revision["directory"], path) if file_info["type"] == "dir": dir_id = file_info["target"] else: query_string = "sha1_git:" + file_info["target"] content_data = request_content(query_string, raise_if_unavailable=False) else: dir_id = revision["directory"] if dir_id: path = "" if path is None else (path + "/") dirs, files = get_directory_entries(dir_id) except Exception as exc: return handle_view_exception(request, exc) revision_metadata = RevisionMetadata( object_type=REVISION, object_id=sha1_git, revision=sha1_git, revision_url=gen_revision_link(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"], directory_url=gen_directory_link(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, snapshot_url=gen_snapshot_link(snapshot_id) if snapshot_id else None, 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_id": None, "revision_context": True, "revision_id": sha1_git, } swh_objects = [SWHObjectInfo(object_type=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 = {} error_code = 200 error_message = "" error_description = "" extra_context = dict(revision_metadata) extra_context["path"] = f"/{path}" if path else "/" 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=CONTENT, object_id=file_info["target"]) ) error_code = content_data["error_code"] error_message = content_data["error_message"] error_description = content_data["error_description"] else: for d in dirs: 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["length"] is not None: f["length"] = filesizeformat(f["length"]) if f["name"].lower().startswith("readme"): readmes[f["name"]] = f["checksums"]["sha1"] readme_name, readme_url, readme_html = get_readme_to_display(readmes) top_right_link = { "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_id"] = dir_id swh_objects.append(SWHObjectInfo(object_type=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=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_menu": True, "swhids_info": swhids_info, "error_code": error_code, "error_message": error_message, "error_description": error_description, }, status=error_code, ) diff --git a/swh/web/common/identifiers.py b/swh/web/common/identifiers.py index 156478b9..085aa154 100644 --- a/swh/web/common/identifiers.py +++ b/swh/web/common/identifiers.py @@ -1,322 +1,394 @@ # Copyright (C) 2020 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 urllib.parse import quote from typing import Any, Dict, Iterable, List, Optional from typing_extensions import TypedDict from django.http import QueryDict from swh.model.exceptions import ValidationError from swh.model.hashutil import hash_to_bytes from swh.model.identifiers import ( persistent_identifier, parse_persistent_identifier, CONTENT, DIRECTORY, ORIGIN, RELEASE, REVISION, SNAPSHOT, PersistentId, ) +from swh.web.common import service from swh.web.common.exc import BadInputExc from swh.web.common.typing import ( QueryParameters, SnapshotContext, SWHObjectInfo, SWHIDInfo, SWHIDContext, ) from swh.web.common.utils import reverse def get_swh_persistent_id( object_type: str, object_id: str, scheme_version: int = 1, metadata: SWHIDContext = {}, ) -> str: """ Returns the persistent identifier for a swh object based on: * the object type * the object id * the swh identifiers scheme version Args: object_type: the swh object type (content/directory/release/revision/snapshot) object_id: the swh object id (hexadecimal representation of its hash value) scheme_version: the scheme version of the swh persistent identifiers Returns: the swh object persistent identifier Raises: BadInputExc: if the provided parameters do not enable to generate a valid identifier """ try: swh_id = persistent_identifier(object_type, object_id, scheme_version, metadata) except ValidationError as e: raise BadInputExc( "Invalid object (%s) for swh persistent id. %s" % (object_id, e) ) else: return swh_id -ResolvedPersistentId = TypedDict( - "ResolvedPersistentId", {"swh_id_parsed": PersistentId, "browse_url": Optional[str]} -) +class ResolvedPersistentId(TypedDict): + """parsed SWHID with context""" + + swh_id_parsed: PersistentId + """URL to browse object according to SWHID context""" + browse_url: Optional[str] def resolve_swh_persistent_id( swh_id: str, query_params: Optional[QueryParameters] = None ) -> ResolvedPersistentId: """ Try to resolve a Software Heritage persistent id into an url for browsing the targeted object. Args: swh_id: a Software Heritage persistent identifier query_params: optional dict filled with query parameters to append to the browse url Returns: a dict with the following keys: * **swh_id_parsed**: the parsed identifier * **browse_url**: the url for browsing the targeted object """ swh_id_parsed = get_persistent_identifier(swh_id) object_type = swh_id_parsed.object_type object_id = swh_id_parsed.object_id browse_url = None + url_args = {} query_dict = QueryDict("", mutable=True) + fragment = "" + anchor_swhid_parsed = None + if query_params and len(query_params) > 0: for k in sorted(query_params.keys()): query_dict[k] = query_params[k] + if "origin" in swh_id_parsed.metadata: query_dict["origin_url"] = swh_id_parsed.metadata["origin"] + + if "anchor" in swh_id_parsed.metadata: + anchor_swhid_parsed = get_persistent_identifier( + swh_id_parsed.metadata["anchor"] + ) + + if "path" in swh_id_parsed.metadata and swh_id_parsed.metadata["path"] != "/": + query_dict["path"] = swh_id_parsed.metadata["path"] + if anchor_swhid_parsed: + directory = "" + if anchor_swhid_parsed.object_type == DIRECTORY: + directory = anchor_swhid_parsed.object_id + elif anchor_swhid_parsed.object_type == REVISION: + revision = service.lookup_revision(anchor_swhid_parsed.object_id) + directory = revision["directory"] + elif anchor_swhid_parsed.object_type == RELEASE: + release = service.lookup_release(anchor_swhid_parsed.object_id) + if release["target_type"] == REVISION: + revision = service.lookup_revision(release["target"]) + directory = revision["directory"] + if object_type == CONTENT: + if "origin" not in swh_id_parsed.metadata: + # when no origin context, content objects need to have their + # path prefixed by root directory id for proper breadcrumbs display + query_dict["path"] = directory + query_dict["path"] + else: + # remove leading slash from SWHID content path + query_dict["path"] = query_dict["path"][1:] + elif object_type == DIRECTORY: + object_id = directory + # remove leading and trailing slashes from SWHID directory path + query_dict["path"] = query_dict["path"][1:-1] + + # snapshot context + if "visit" in swh_id_parsed.metadata: + + snp_swhid_parsed = get_persistent_identifier(swh_id_parsed.metadata["visit"]) + if snp_swhid_parsed.object_type != SNAPSHOT: + raise BadInputExc("Visit must be a snapshot SWHID.") + query_dict["snapshot"] = snp_swhid_parsed.object_id + + if anchor_swhid_parsed: + if anchor_swhid_parsed.object_type == REVISION: + # check if the anchor revision is the tip of a branch + branch_name = service.lookup_snapshot_branch_name_from_tip_revision( + snp_swhid_parsed.object_id, anchor_swhid_parsed.object_id + ) + if branch_name: + query_dict["branch"] = branch_name + elif object_type != REVISION: + query_dict["revision"] = anchor_swhid_parsed.object_id + + elif anchor_swhid_parsed.object_type == RELEASE: + release = service.lookup_release(anchor_swhid_parsed.object_id) + if release: + query_dict["release"] = release["name"] + + if object_type == REVISION and "release" not in query_dict: + branch_name = service.lookup_snapshot_branch_name_from_tip_revision( + snp_swhid_parsed.object_id, object_id + ) + if branch_name: + query_dict["branch"] = branch_name + + # browsing content or directory without snapshot context + elif object_type in (CONTENT, DIRECTORY) and anchor_swhid_parsed: + if anchor_swhid_parsed.object_type == REVISION: + # anchor revision, objects are browsed from its view + object_type = REVISION + object_id = anchor_swhid_parsed.object_id + elif object_type == DIRECTORY and anchor_swhid_parsed.object_type == DIRECTORY: + # a directory is browsed from its root + object_id = anchor_swhid_parsed.object_id + if object_type == CONTENT: query_string = "sha1_git:" + object_id - fragment = "" if "lines" in swh_id_parsed.metadata: lines = swh_id_parsed.metadata["lines"].split("-") fragment += "#L" + lines[0] if len(lines) > 1: fragment += "-L" + lines[1] - browse_url = ( - reverse( - "browse-content", - url_args={"query_string": query_string}, - query_params=query_dict, - ) - + fragment - ) + url_args["query_string"] = query_string + elif object_type == DIRECTORY: - browse_url = reverse( - "browse-directory", - url_args={"sha1_git": object_id}, - query_params=query_dict, - ) + url_args["sha1_git"] = object_id elif object_type == RELEASE: - browse_url = reverse( - "browse-release", url_args={"sha1_git": object_id}, query_params=query_dict - ) + url_args["sha1_git"] = object_id elif object_type == REVISION: - browse_url = reverse( - "browse-revision", url_args={"sha1_git": object_id}, query_params=query_dict - ) + url_args["sha1_git"] = object_id elif object_type == SNAPSHOT: - browse_url = reverse( - "browse-snapshot", - url_args={"snapshot_id": object_id}, - query_params=query_dict, - ) + url_args["snapshot_id"] = object_id elif object_type == ORIGIN: raise BadInputExc( ( "Origin PIDs (Persistent Identifiers) are not " "publicly resolvable because they are for " "internal usage only" ) ) - return {"swh_id_parsed": swh_id_parsed, "browse_url": browse_url} + if url_args: + browse_url = ( + reverse( + f"browse-{object_type}", url_args=url_args, query_params=query_dict, + ) + + fragment + ) + + return ResolvedPersistentId(swh_id_parsed=swh_id_parsed, browse_url=browse_url) def get_persistent_identifier(persistent_id: str) -> PersistentId: """Check if a persistent identifier is valid. Args: persistent_id: A string representing a Software Heritage persistent identifier. Raises: BadInputExc: if the provided persistent identifier can not be parsed. Return: A persistent identifier object. """ try: pid_object = parse_persistent_identifier(persistent_id) except ValidationError as ve: raise BadInputExc("Error when parsing identifier: %s" % " ".join(ve.messages)) else: return pid_object def group_swh_persistent_identifiers( persistent_ids: Iterable[PersistentId], ) -> Dict[str, List[bytes]]: """ Groups many Software Heritage persistent identifiers into a dictionary depending on their type. Args: persistent_ids: an iterable of Software Heritage persistent identifier objects Returns: A dictionary with: keys: persistent identifier types values: persistent identifiers id """ pids_by_type: Dict[str, List[bytes]] = { CONTENT: [], DIRECTORY: [], REVISION: [], RELEASE: [], SNAPSHOT: [], } for pid in persistent_ids: obj_id = pid.object_id obj_type = pid.object_type pids_by_type[obj_type].append(hash_to_bytes(obj_id)) return pids_by_type def get_swhids_info( swh_objects: Iterable[SWHObjectInfo], snapshot_context: Optional[SnapshotContext] = None, extra_context: Optional[Dict[str, Any]] = None, ) -> List[SWHIDInfo]: """ Returns a list of dict containing info related to persistent identifiers of swh objects. Args: swh_objects: an iterable of dict describing archived objects snapshot_context: optional dict parameter describing the snapshot in which the objects have been found extra_context: optional dict filled with extra contextual info about the objects Returns: a list of dict containing persistent identifiers info """ swhids_info = [] for swh_object in swh_objects: if not swh_object["object_id"]: swhids_info.append( SWHIDInfo( object_type=swh_object["object_type"], object_id="", swhid="", swhid_url="", context={}, swhid_with_context=None, swhid_with_context_url=None, ) ) continue object_type = swh_object["object_type"] object_id = swh_object["object_id"] swhid_context: SWHIDContext = {} if snapshot_context: if snapshot_context["origin_info"] is not None: swhid_context["origin"] = quote( snapshot_context["origin_info"]["url"], safe="/?:@&" ) if object_type != SNAPSHOT: swhid_context["visit"] = get_swh_persistent_id( SNAPSHOT, snapshot_context["snapshot_id"] ) if object_type in (CONTENT, DIRECTORY): if snapshot_context["release_id"] is not None: swhid_context["anchor"] = get_swh_persistent_id( RELEASE, snapshot_context["release_id"] ) elif snapshot_context["revision_id"] is not None: swhid_context["anchor"] = get_swh_persistent_id( REVISION, snapshot_context["revision_id"] ) if object_type in (CONTENT, DIRECTORY): if ( extra_context and "revision" in extra_context and extra_context["revision"] and "anchor" not in swhid_context ): swhid_context["anchor"] = get_swh_persistent_id( REVISION, extra_context["revision"] ) elif ( extra_context and "root_directory" in extra_context and extra_context["root_directory"] and "anchor" not in swhid_context and ( object_type != DIRECTORY or extra_context["root_directory"] != object_id ) ): swhid_context["anchor"] = get_swh_persistent_id( DIRECTORY, extra_context["root_directory"] ) path = None if extra_context and "path" in extra_context: path = extra_context["path"] or "/" if "filename" in extra_context and object_type == CONTENT: path += extra_context["filename"] if path: swhid_context["path"] = quote(path, safe="/?:@&") swhid = get_swh_persistent_id(object_type, object_id) swhid_url = reverse("browse-swh-id", url_args={"swh_id": swhid}) swhid_with_context = None swhid_with_context_url = None if swhid_context: swhid_with_context = get_swh_persistent_id( object_type, object_id, metadata=swhid_context ) swhid_with_context_url = reverse( "browse-swh-id", url_args={"swh_id": swhid_with_context} ) swhids_info.append( SWHIDInfo( object_type=object_type, object_id=object_id, swhid=swhid, swhid_url=swhid_url, context=swhid_context, swhid_with_context=swhid_with_context, swhid_with_context_url=swhid_with_context_url, ) ) return swhids_info diff --git a/swh/web/common/service.py b/swh/web/common/service.py index b87c0ab3..b201019a 100644 --- a/swh/web/common/service.py +++ b/swh/web/common/service.py @@ -1,1248 +1,1283 @@ # Copyright (C) 2015-2020 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 itertools import os import re from collections import defaultdict from typing import Any, Dict, List, Set, Iterator, Optional, Tuple from swh.model import hashutil from swh.model.identifiers import CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT from swh.storage.algos import diff, revisions_walker from swh.vault.exc import NotFoundExc as VaultNotFoundExc from swh.web import config from swh.web.common import converters from swh.web.common import query from swh.web.common.exc import BadInputExc, NotFoundExc from swh.web.common.origin_visits import get_origin_visit from swh.web.common.typing import OriginInfo, OriginVisitInfo search = config.search() storage = config.storage() vault = config.vault() idx_storage = config.indexer_storage() 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): """Checks if the storage contains a given content checksum Args: 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})) return {"found": converters.from_content(found), "algo": algo} def search_hash(q): """Checks if the storage contains a given content checksum Args: 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): """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) def lookup_content_language(q): """Return language information from a specified content. Args: q: query string of the form Yields: language information (dict) list if the content is found. """ sha1 = _lookup_content_sha1(q) if not sha1: return None lang = _first_element(list(idx_storage.content_language_get([sha1]))) if not lang: return None return converters.from_swh(lang, hashess={"id"}) 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 lic = _first_element(idx_storage.content_fossology_license_get([sha1])) if not lic: return None return converters.from_swh({"id": sha1, "facts": lic[sha1]}, 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. """ origins = [origin] 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("/"): origins.append({"url": 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: origins.append({"url": f"{origin['url']}/"}) # Check all possible origin urls for orig in origins: origin_info = storage.origin_get(orig) if origin_info: break if not origin_info: msg = "Origin with url %s not found!" % origin["url"] raise NotFoundExc(msg) return converters.from_origin(origin_info) def lookup_origins( origin_from: int = 1, origin_count: int = 100 ) -> Iterator[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 Yields: origins information as dicts """ origins = storage.origin_get_range(origin_from, origin_count) return map(converters.from_origin, origins) def search_origin( url_pattern: str, limit: int = 50, with_visit: bool = False, page_token: Any = None ) -> Tuple[List[OriginInfo], Any]: """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 limit: the maximum number of found origins to return page_token: opaque string used to get the next results of a search Returns: list of origin information as dict. """ if search: results = search.origin_search( url_pattern=url_pattern, count=limit, page_token=page_token, with_visit=with_visit, ) origins = list(map(converters.from_origin, results["results"])) return (origins, results["next_page_token"]) else: # Fallback to swh-storage if swh-search is not configured offset = int(page_token) if page_token else 0 regexp = True 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) origins_raw = storage.origin_search( url_pattern, offset, limit, regexp, with_visit ) origins = list(map(converters.from_origin, origins_raw)) if len(origins) >= limit: page_token = str(offset + len(origins)) else: page_token = None return (origins, page_token) def search_origin_metadata(fulltext, limit=50): """Search for origins whose metadata match a provided string pattern. Args: fulltext: the string pattern to search for in origin metadata offset: number of found origins to skip before returning results limit: the maximum number of found origins to return Returns: list of origin metadata as dict. """ matches = idx_storage.origin_intrinsic_metadata_search_fulltext( conjunction=[fulltext], limit=limit ) results = [] for match in matches: match["from_revision"] = hashutil.hash_to_hex(match["from_revision"]) origin = storage.origin_get({"url": match["id"]}) del match["id"] result = converters.from_origin(origin) if result: result["metadata"] = match results.append(result) return results def lookup_origin_intrinsic_metadata(origin_dict): """Return intrinsic metadata for origin whose origin matches given origin. Args: origin_dict: origin's dict with keys ('type' AND 'url') Returns: origin metadata. """ origin_info = storage.origin_get(origin_dict) if not origin_info: msg = "Origin with url %s not found!" % origin_dict["url"] raise NotFoundExc(msg) origins = [origin_info["url"]] 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, path_string): """Return directory information for entry with path path_string w.r.t. root directory pointed by directory_sha1_git Args: - directory_sha1_git: sha1_git corresponding to the directory to which we append paths to (hopefully) find the entry - the relative path to the entry starting from the directory pointed by directory_sha1_git 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_string.strip(os.path.sep).split(os.path.sep) queried_dir = storage.directory_entry_get_by_path( sha1_git_bin, list(map(lambda p: p.encode("utf-8"), paths)) ) if not queried_dir: raise NotFoundExc( ("Directory entry with path %s from %s not found") % (path_string, sha1_git) ) return converters.from_directory_entry(queried_dir) def lookup_release(release_sha1_git): """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("Release with sha1_git %s not found." % release_sha1_git) return converters.from_release(release) def lookup_release_multiple(sha1_git_list): """Return information about the revisions identified with their sha1_git identifiers. Args: sha1_git_list: A list of revision sha1_git identifiers Returns: Release 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) or [] return (converters.from_release(r) for r in releases) def lookup_revision(rev_sha1_git): """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 = _first_element(storage.revision_get([sha1_git_bin])) if not revision: raise NotFoundExc("Revision with sha1_git %s not found." % rev_sha1_git) return converters.from_revision(revision) def lookup_revision_multiple(sha1_git_list): """Return information about the revisions identified with their sha1_git identifiers. Args: sha1_git_list: A list of revision sha1_git identifiers Returns: Iterator of revisions 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) revisions = storage.revision_get(sha1_bin_list) or [] return (converters.from_revision(r) for r in revisions) def lookup_revision_message(rev_sha1_git): """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 = _first_element(storage.revision_get([sha1_git_bin])) if not revision: raise NotFoundExc("Revision with sha1_git %s not found." % rev_sha1_git) if "message" not in revision: raise NotFoundExc("No message for revision with sha1_git %s." % rev_sha1_git) res = {"message": revision["message"]} return res 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 = 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.') 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) 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 = _first_element(storage.revision_get([rev_root_id_bin])) return ( converters.from_revision(rev_root), lookup_revision_with_context(rev_root, sha1_git, limit), ) def lookup_revision_with_context(sha1_git_root, sha1_git, limit=100): """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 = _first_element(storage.revision_get([sha1_git_bin])) if not revision: raise NotFoundExc("Revision %s not found" % sha1_git) if isinstance(sha1_git_root, str): sha1_git_root_bin = _to_sha1_bin(sha1_git_root) revision_root = _first_element(storage.revision_get([sha1_git_root_bin])) if not revision_root: raise NotFoundExc("Revision root %s not found" % sha1_git_root) else: sha1_git_root_bin = sha1_git_root["id"] revision_log = storage.revision_log([sha1_git_root_bin], limit) parents = {} 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( "Revision %s is not an ancestor of %s" % (sha1_git, sha1_git_root) ) revision["children"] = children[revision["id"]] return converters.from_revision(revision) 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 = _first_element(storage.revision_get([sha1_git_bin])) if not revision: raise NotFoundExc("Revision %s not found" % sha1_git) 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("Content not found for revision %s" % sha1_git) if with_data: c = _first_element(storage.content_get([content["sha1"]])) content["data"] = c["data"] return { "type": "file", "path": "." if not dir_path else dir_path, "revision": sha1_git, "content": converters.from_content(content), } elif entity["type"] == "rev": # revision revision = next(storage.revision_get([entity["target"]])) return { "type": "rev", "path": "." if not dir_path else dir_path, "revision": sha1_git, "content": converters.from_revision(revision), } else: raise NotImplementedError("Entity of type %s not implemented." % entity["type"]) def lookup_content(q): """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: raise NotFoundExc( "Content with %s checksum equals to %s not found!" % (algo, hashutil.hash_to_hex(hash)) ) return converters.from_content(c) def lookup_content_raw(q): """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 = _first_element(storage.content_get([content_sha1_bytes])) if not content: algo, hash = query.parse_hash(q) raise NotFoundExc( "Bytes of content with %s checksum equals to %s " "are not available!" % (algo, hashutil.hash_to_hex(hash)) ) return converters.from_content(content) def stat_counters(): """Return the stat counters for Software Heritage Returns: A dict mapping textual labels to integer values. """ return storage.stat_counters() def _lookup_origin_visits( origin_url: str, last_visit: Optional[int] = None, limit: int = 10 ) -> Iterator[Dict[str, Any]]: """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: Dictionaries of origin_visit for that origin """ limit = min(limit, MAX_LIMIT) for visit in storage.origin_visit_get( origin_url, last_visit=last_visit, limit=limit ): visit["origin"] = origin_url yield visit 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 """ visits = _lookup_origin_visits(origin, last_visit=last_visit, limit=per_page) for visit in visits: yield converters.from_origin_visit(visit) def lookup_origin_visit_latest( origin_url: str, require_snapshot: bool ) -> OriginVisitInfo: """Return the origin's latest visit Args: origin_url (str): origin to list visits for require_snapshot (bool): filter out origins without a snapshot Returns: dict: The origin_visit concerned """ visit = storage.origin_visit_get_latest( origin_url, require_snapshot=require_snapshot ) return converters.from_origin_visit(visit) def lookup_origin_visit(origin_url: str, visit_id: int) -> OriginVisitInfo: """Return information about visit visit_id with origin origin. Args: origin (str): origin concerned by the visit visit_id: the visit identifier to lookup Yields: The dict origin_visit concerned """ visit = storage.origin_visit_get_by(origin_url, visit_id) if not visit: raise NotFoundExc( "Origin %s or its visit " "with id %s not found!" % (origin_url, visit_id) ) visit["origin"] = origin_url return converters.from_origin_visit(visit) def lookup_snapshot_sizes(snapshot_id): """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 = storage.snapshot_count_branches(snapshot_id_bin) if "revision" not in snapshot_sizes: snapshot_sizes["revision"] = 0 if "release" not in snapshot_sizes: snapshot_sizes["release"] = 0 # adjust revision / release count for display if aliases are defined if "alias" in snapshot_sizes: aliases = lookup_snapshot( snapshot_id, branches_count=snapshot_sizes["alias"], target_types=["alias"] ) for alias in aliases["branches"].values(): if lookup_snapshot( snapshot_id, branches_from=alias["target"], branches_count=1, target_types=["revision"], ): snapshot_sizes["revision"] += 1 else: snapshot_sizes["release"] += 1 del snapshot_sizes["alias"] return snapshot_sizes def lookup_snapshot( snapshot_id, branches_from="", branches_count=1000, target_types=None ): """Return information about a snapshot, aka the list of named branches found during a specific visit of an origin. Args: snapshot_id (str): sha1 identifier of the snapshot branches_from (str): optional parameter used to skip branches whose name is lesser than it before returning them branches_count (int): optional parameter used to restrain the amount of returned branches target_types (list): 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'`) Returns: A dict filled with the snapshot content. """ snapshot_id_bin = _to_sha1_bin(snapshot_id) snapshot = storage.snapshot_get_branches( snapshot_id_bin, branches_from.encode(), branches_count, target_types ) if not snapshot: raise NotFoundExc("Snapshot with id %s not found!" % snapshot_id) return converters.from_snapshot(snapshot) def lookup_latest_origin_snapshot(origin, allowed_statuses=None): """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. """ snapshot = storage.snapshot_get_latest(origin, allowed_statuses) return converters.from_snapshot(snapshot) +def lookup_snapshot_branch_name_from_tip_revision( + snapshot_id: str, revision_id: str +) -> Optional[str]: + """Check if a revision corresponds to the tip of a snapshot branch + + Args: + snapshot_id: hexadecimal representation of a snapshot id + revision_id: hexadecimal representation of a revision id + + Returns: + The name of the first found branch or None otherwise + """ + per_page = 10000 + branches_from = "" + snapshot: Dict[str, Any] = {"branches": {}} + branches = [] + while not branches_from or len(snapshot["branches"]) == per_page + 1: + snapshot = lookup_snapshot( + snapshot_id, + target_types=[REVISION], + branches_from=branches_from, + branches_count=per_page + 1, + ) + + branches += [ + {"name": k, "target": v["target"]} for k, v in snapshot["branches"].items() + ] + branches_from = branches[-1]["name"] + + for branch in branches: + if branch["target"] == revision_id: + return branch["name"] + return 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, *args, **kwargs): try: return vault_fn(*args, **kwargs) except VaultNotFoundExc: return None def vault_cook(obj_type, obj_id, email=None): """Cook a vault bundle. """ return _vault_request(vault.cook, obj_type, obj_id, email=email) def vault_fetch(obj_type, obj_id): """Fetch a vault bundle. """ return _vault_request(vault.fetch, obj_type, obj_id) def vault_progress(obj_type, obj_id): """Get the current progress of a vault bundle. """ return _vault_request(vault.progress, obj_type, obj_id) 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: str, 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 == CONTENT: return lookup_content(f"sha1_git:{object_id}") elif object_type == DIRECTORY: return {"id": object_id, "content": list(lookup_directory(object_id))} elif object_type == RELEASE: return lookup_release(object_id) elif object_type == REVISION: return lookup_revision(object_id) elif object_type == SNAPSHOT: return lookup_snapshot(object_id) raise BadInputExc( ( "Invalid swh object type! Valid types are " f"{CONTENT}, {DIRECTORY}, {RELEASE} " f"{REVISION} or {SNAPSHOT}." ) ) def lookup_missing_hashes(grouped_pids: Dict[str, List[bytes]]) -> Set[str]: """Lookup missing Software Heritage persistent identifier hash, using batch processing. Args: A dictionary with: keys: persistent identifier type values: list(bytes) persistent identifier hash Returns: A set(hexadecimal) of the hashes not found in the storage """ missing_hashes = [] for obj_type, obj_ids in grouped_pids.items(): if obj_type == CONTENT: missing_hashes.append(storage.content_missing_per_sha1_git(obj_ids)) elif obj_type == DIRECTORY: missing_hashes.append(storage.directory_missing(obj_ids)) elif obj_type == REVISION: missing_hashes.append(storage.revision_missing(obj_ids)) elif obj_type == RELEASE: missing_hashes.append(storage.directory_missing(obj_ids)) elif obj_type == SNAPSHOT: missing_hashes.append(storage.directory_missing(obj_ids)) missing = set( map(lambda x: hashutil.hash_to_hex(x), itertools.chain(*missing_hashes)) ) return missing diff --git a/swh/web/tests/common/test_identifiers.py b/swh/web/tests/common/test_identifiers.py index dbe3cf6c..788b453c 100644 --- a/swh/web/tests/common/test_identifiers.py +++ b/swh/web/tests/common/test_identifiers.py @@ -1,419 +1,566 @@ # Copyright (C) 2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import random from hypothesis import given import pytest from swh.model.hashutil import hash_to_bytes from swh.model.identifiers import ( CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT, PersistentId, ) from swh.web.common.exc import BadInputExc from swh.web.common.identifiers import ( get_swh_persistent_id, resolve_swh_persistent_id, get_persistent_identifier, group_swh_persistent_identifiers, get_swhids_info, ) from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.common.utils import reverse from swh.web.common.typing import SWHObjectInfo from swh.web.tests.data import random_sha1 from swh.web.tests.strategies import ( content, directory, release, revision, snapshot, origin, origin_with_multiple_visits, directory_with_subdirs, ) @given(content()) def test_get_swh_persistent_id(content): swh_object_type = CONTENT sha1_git = content["sha1_git"] expected_swh_id = "swh:1:cnt:" + sha1_git assert get_swh_persistent_id(swh_object_type, sha1_git) == expected_swh_id with pytest.raises(BadInputExc) as e: get_swh_persistent_id("foo", sha1_git) assert e.match("Invalid object") with pytest.raises(BadInputExc) as e: get_swh_persistent_id(swh_object_type, "not a valid id") assert e.match("Invalid object") @given(content(), directory(), release(), revision(), snapshot()) -def test_resolve_swh_persistent_id(content, directory, release, revision, snapshot): +def test_resolve_swh_persistent_id_legacy( + content, directory, release, revision, snapshot +): for obj_type, obj_id in ( (CONTENT, content["sha1_git"]), (DIRECTORY, directory), (RELEASE, release), (REVISION, revision), (SNAPSHOT, snapshot), ): swh_pid = get_swh_persistent_id(obj_type, obj_id) url_args = {} if obj_type == CONTENT: url_args["query_string"] = f"sha1_git:{obj_id}" elif obj_type == SNAPSHOT: url_args["snapshot_id"] = obj_id else: url_args["sha1_git"] = obj_id query_params = {"origin_url": "some-origin"} browse_url = reverse( f"browse-{obj_type}", url_args=url_args, query_params=query_params ) resolved_pid = resolve_swh_persistent_id(swh_pid, query_params) assert isinstance(resolved_pid["swh_id_parsed"], PersistentId) assert str(resolved_pid["swh_id_parsed"]) == swh_pid assert resolved_pid["browse_url"] == browse_url with pytest.raises(BadInputExc, match="Origin PIDs"): resolve_swh_persistent_id(f"swh:1:ori:{random_sha1()}") @given(content(), directory(), release(), revision(), snapshot()) def test_get_persistent_identifier(content, directory, release, revision, snapshot): for obj_type, obj_id in ( (CONTENT, content["sha1_git"]), (DIRECTORY, directory), (RELEASE, release), (REVISION, revision), (SNAPSHOT, snapshot), ): swh_pid = get_swh_persistent_id(obj_type, obj_id) swh_parsed_pid = get_persistent_identifier(swh_pid) assert isinstance(swh_parsed_pid, PersistentId) assert str(swh_parsed_pid) == swh_pid with pytest.raises(BadInputExc, match="Error when parsing identifier"): get_persistent_identifier("foo") @given(content(), directory(), release(), revision(), snapshot()) def test_group_persistent_identifiers(content, directory, release, revision, snapshot): swh_pids = [] expected = {} for obj_type, obj_id in ( (CONTENT, content["sha1_git"]), (DIRECTORY, directory), (RELEASE, release), (REVISION, revision), (SNAPSHOT, snapshot), ): swh_pid = get_swh_persistent_id(obj_type, obj_id) swh_pid = get_persistent_identifier(swh_pid) swh_pids.append(swh_pid) expected[obj_type] = [hash_to_bytes(obj_id)] pid_groups = group_swh_persistent_identifiers(swh_pids) assert pid_groups == expected @given(directory_with_subdirs()) def test_get_swhids_info_directory_context(archive_data, directory): extra_context = {"path": "/"} swhid = get_swhids_info( [SWHObjectInfo(object_type=DIRECTORY, object_id=directory)], snapshot_context=None, extra_context=extra_context, )[0] swhid_dir_parsed = get_persistent_identifier(swhid["swhid_with_context"]) assert swhid_dir_parsed.metadata == extra_context dir_content = archive_data.directory_ls(directory) dir_subdirs = [e for e in dir_content if e["type"] == "dir"] dir_subdir = random.choice(dir_subdirs) dir_subdir_path = f'/{dir_subdir["name"]}/' dir_subdir_content = archive_data.directory_ls(dir_subdir["target"]) dir_subdir_files = [e for e in dir_subdir_content if e["type"] == "file"] dir_subdir_file = random.choice(dir_subdir_files) extra_context = { "root_directory": directory, "path": dir_subdir_path, "filename": dir_subdir_file["name"], } swhids = get_swhids_info( [ SWHObjectInfo(object_type=DIRECTORY, object_id=dir_subdir["target"]), SWHObjectInfo( object_type=CONTENT, object_id=dir_subdir_file["checksums"]["sha1_git"] ), ], snapshot_context=None, extra_context=extra_context, ) swhid_dir_parsed = get_persistent_identifier(swhids[0]["swhid_with_context"]) swhid_cnt_parsed = get_persistent_identifier(swhids[1]["swhid_with_context"]) anchor = get_swh_persistent_id(DIRECTORY, directory) assert swhid_dir_parsed.metadata == { "anchor": anchor, "path": dir_subdir_path, } assert swhid_cnt_parsed.metadata == { "anchor": anchor, "path": f'{dir_subdir_path}{dir_subdir_file["name"]}', } @given(revision()) def test_get_swhids_info_revision_context(archive_data, revision): revision_data = archive_data.revision_get(revision) directory = revision_data["directory"] dir_content = archive_data.directory_ls(directory) dir_entry = random.choice(dir_content) swh_objects = [ SWHObjectInfo(object_type=REVISION, object_id=revision), SWHObjectInfo(object_type=DIRECTORY, object_id=directory), ] extra_context = {"revision": revision, "path": "/"} if dir_entry["type"] == "file": swh_objects.append( SWHObjectInfo( object_type=CONTENT, object_id=dir_entry["checksums"]["sha1_git"] ) ) extra_context["filename"] = dir_entry["name"] swhids = get_swhids_info( swh_objects, snapshot_context=None, extra_context=extra_context, ) assert swhids[0]["context"] == {} swhid_dir_parsed = get_persistent_identifier(swhids[1]["swhid_with_context"]) anchor = get_swh_persistent_id(REVISION, revision) assert swhid_dir_parsed.metadata == { "anchor": anchor, "path": "/", } if dir_entry["type"] == "file": swhid_cnt_parsed = get_persistent_identifier(swhids[2]["swhid_with_context"]) assert swhid_cnt_parsed.metadata == { "anchor": anchor, "path": f'/{dir_entry["name"]}', } @given(origin_with_multiple_visits()) def test_get_swhids_info_origin_snapshot_context(archive_data, origin): """ Test SWHIDs with contextual info computation under a variety of origin / snapshot browsing contexts. """ visits = archive_data.origin_visit_get(origin["url"]) for visit in visits: snapshot = archive_data.snapshot_get(visit["snapshot"]) snapshot_id = snapshot["id"] branches = { k: v["target"] for k, v in snapshot["branches"].items() if v["target_type"] == "revision" } releases = { k: v["target"] for k, v in snapshot["branches"].items() if v["target_type"] == "release" } head_rev_id = archive_data.snapshot_get_head(snapshot) head_rev = archive_data.revision_get(head_rev_id) root_dir = head_rev["directory"] dir_content = archive_data.directory_ls(root_dir) dir_files = [e for e in dir_content if e["type"] == "file"] dir_file = random.choice(dir_files) revision_log = [r["id"] for r in archive_data.revision_log(head_rev_id)] branch_name = random.choice(list(branches)) release = random.choice(list(releases)) release_data = archive_data.release_get(releases[release]) release_name = release_data["name"] revision_id = random.choice(revision_log) for snp_ctx_params, anchor_info in ( ( {"snapshot_id": snapshot_id}, {"anchor_type": REVISION, "anchor_id": head_rev_id}, ), ( {"snapshot_id": snapshot_id, "branch_name": branch_name}, {"anchor_type": REVISION, "anchor_id": branches[branch_name]}, ), ( {"snapshot_id": snapshot_id, "release_name": release_name}, {"anchor_type": RELEASE, "anchor_id": releases[release]}, ), ( {"snapshot_id": snapshot_id, "revision_id": revision_id}, {"anchor_type": REVISION, "anchor_id": revision_id}, ), ( {"origin_url": origin["url"], "snapshot_id": snapshot_id}, {"anchor_type": REVISION, "anchor_id": head_rev_id}, ), ( { "origin_url": origin["url"], "snapshot_id": snapshot_id, "branch_name": branch_name, }, {"anchor_type": REVISION, "anchor_id": branches[branch_name]}, ), ( { "origin_url": origin["url"], "snapshot_id": snapshot_id, "release_name": release_name, }, {"anchor_type": RELEASE, "anchor_id": releases[release]}, ), ( { "origin_url": origin["url"], "snapshot_id": snapshot_id, "revision_id": revision_id, }, {"anchor_type": REVISION, "anchor_id": revision_id}, ), ): snapshot_context = get_snapshot_context(**snp_ctx_params) rev_id = head_rev_id if "branch_name" in snp_ctx_params: rev_id = branches[branch_name] elif "release_name" in snp_ctx_params: rev_id = release_data["target"] elif "revision_id" in snp_ctx_params: rev_id = revision_id swh_objects = [ SWHObjectInfo( object_type=CONTENT, object_id=dir_file["checksums"]["sha1_git"] ), SWHObjectInfo(object_type=DIRECTORY, object_id=root_dir), SWHObjectInfo(object_type=REVISION, object_id=rev_id), SWHObjectInfo(object_type=SNAPSHOT, object_id=snapshot_id), ] if "release_name" in snp_ctx_params: swh_objects.append( SWHObjectInfo(object_type=RELEASE, object_id=release_data["id"]) ) swhids = get_swhids_info( swh_objects, snapshot_context, extra_context={"path": "/", "filename": dir_file["name"]}, ) swhid_cnt_parsed = get_persistent_identifier( swhids[0]["swhid_with_context"] ) swhid_dir_parsed = get_persistent_identifier( swhids[1]["swhid_with_context"] ) swhid_rev_parsed = get_persistent_identifier( swhids[2]["swhid_with_context"] ) swhid_snp_parsed = get_persistent_identifier( swhids[3]["swhid_with_context"] or swhids[3]["swhid"] ) swhid_rel_parsed = None if "release_name" in snp_ctx_params: swhid_rel_parsed = get_persistent_identifier( swhids[4]["swhid_with_context"] ) anchor = get_swh_persistent_id( object_type=anchor_info["anchor_type"], object_id=anchor_info["anchor_id"], ) snapshot_swhid = get_swh_persistent_id( object_type=SNAPSHOT, object_id=snapshot_id ) expected_cnt_context = { "visit": snapshot_swhid, "anchor": anchor, "path": f'/{dir_file["name"]}', } expected_dir_context = { "visit": snapshot_swhid, "anchor": anchor, "path": "/", } expected_rev_context = {"visit": snapshot_swhid} expected_snp_context = {} if "origin_url" in snp_ctx_params: expected_cnt_context["origin"] = origin["url"] expected_dir_context["origin"] = origin["url"] expected_rev_context["origin"] = origin["url"] expected_snp_context["origin"] = origin["url"] assert swhid_cnt_parsed.metadata == expected_cnt_context assert swhid_dir_parsed.metadata == expected_dir_context assert swhid_rev_parsed.metadata == expected_rev_context assert swhid_snp_parsed.metadata == expected_snp_context if "release_name" in snp_ctx_params: assert swhid_rel_parsed.metadata == expected_rev_context @given(origin(), directory()) def test_get_swhids_info_path_encoding(archive_data, origin, directory): snapshot_context = get_snapshot_context(origin_url=origin["url"]) snapshot_context["origin_info"]["url"] = "http://example.org/?project=abc;def%" path = "/foo;/bar%" swhid = get_swhids_info( [SWHObjectInfo(object_type=DIRECTORY, object_id=directory)], snapshot_context=snapshot_context, extra_context={"path": path}, )[0] assert swhid["context"]["origin"] == "http://example.org/?project%3Dabc%3Bdef%25" assert swhid["context"]["path"] == "/foo%3B/bar%25" + + +@given(origin_with_multiple_visits()) +def test_resolve_swhids_snapshot_context(client, archive_data, origin): + visits = archive_data.origin_visit_get(origin["url"]) + visit = random.choice(visits) + snapshot = archive_data.snapshot_get(visit["snapshot"]) + head_rev_id = archive_data.snapshot_get_head(snapshot) + branch_info = random.choice( + [ + {"name": k, "revision": v["target"]} + for k, v in snapshot["branches"].items() + if v["target_type"] == "revision" + ] + ) + release_info = random.choice( + [ + {"name": k, "release": v["target"]} + for k, v in snapshot["branches"].items() + if v["target_type"] == "release" + ] + ) + release_info["name"] = archive_data.release_get(release_info["release"])["name"] + + directory = archive_data.revision_get(branch_info["revision"])["directory"] + directory_content = archive_data.directory_ls(directory) + directory_subdir = random.choice( + [e for e in directory_content if e["type"] == "dir"] + ) + directory_file = random.choice( + [e for e in directory_content if e["type"] == "file"] + ) + random_rev_id = random.choice(archive_data.revision_log(head_rev_id))["id"] + + for snp_ctx_params in ( + {}, + {"branch_name": branch_info["name"]}, + {"release_name": release_info["name"]}, + {"revision_id": random_rev_id}, + ): + snapshot_context = get_snapshot_context( + snapshot["id"], origin["url"], **snp_ctx_params + ) + + _check_resolved_swhid_browse_url(SNAPSHOT, snapshot["id"], snapshot_context) + + rev = head_rev_id + if "branch_name" in snp_ctx_params: + rev = branch_info["revision"] + if "revision_id" in snp_ctx_params: + rev = random_rev_id + + _check_resolved_swhid_browse_url(REVISION, rev, snapshot_context) + + _check_resolved_swhid_browse_url( + DIRECTORY, directory, snapshot_context, path="/" + ) + + _check_resolved_swhid_browse_url( + DIRECTORY, + directory_subdir["target"], + snapshot_context, + path=f"/{directory_subdir['name']}/", + ) + + _check_resolved_swhid_browse_url( + CONTENT, + directory_file["target"], + snapshot_context, + path=f"/{directory_file['name']}", + ) + + +def _check_resolved_swhid_browse_url( + object_type, object_id, snapshot_context, path=None +): + snapshot_id = snapshot_context["snapshot_id"] + origin_url = None + if snapshot_context["origin_info"]: + origin_url = snapshot_context["origin_info"]["url"] + + obj_context = {} + query_params = {} + + if origin_url: + obj_context["origin"] = origin_url + query_params["origin_url"] = origin_url + + obj_context["visit"] = get_swh_persistent_id(SNAPSHOT, snapshot_id) + query_params["snapshot"] = snapshot_id + + if object_type in (CONTENT, DIRECTORY, REVISION): + if snapshot_context["release"]: + obj_context["anchor"] = get_swh_persistent_id( + RELEASE, snapshot_context["release_id"] + ) + query_params["release"] = snapshot_context["release"] + else: + obj_context["anchor"] = get_swh_persistent_id( + REVISION, snapshot_context["revision_id"] + ) + if ( + snapshot_context["branch"] + and snapshot_context["branch"] != snapshot_context["revision_id"] + ): + branch = snapshot_context["branch"] + if branch == "HEAD": + for b in snapshot_context["branches"]: + if ( + b["revision"] == snapshot_context["revision_id"] + and b["name"] != "HEAD" + ): + branch = b["name"] + break + + query_params["branch"] = branch + elif object_type != REVISION: + query_params["revision"] = snapshot_context["revision_id"] + + if path: + obj_context["path"] = path + if path != "/": + if object_type == CONTENT: + query_params["path"] = path[1:] + else: + query_params["path"] = path[1:-1] + + if object_type == DIRECTORY: + object_id = snapshot_context["root_directory"] + + obj_swhid = get_swh_persistent_id(object_type, object_id, metadata=obj_context) + + obj_swhid_resolved = resolve_swh_persistent_id(obj_swhid) + + url_args = {"sha1_git": object_id} + if object_type == CONTENT: + url_args = {"query_string": f"sha1_git:{object_id}"} + elif object_type == SNAPSHOT: + url_args = {"snapshot_id": object_id} + + expected_url = reverse( + f"browse-{object_type}", url_args=url_args, query_params=query_params, + ) + + assert obj_swhid_resolved["browse_url"] == expected_url diff --git a/swh/web/tests/common/test_service.py b/swh/web/tests/common/test_service.py index 33fd4df8..778bc45b 100644 --- a/swh/web/tests/common/test_service.py +++ b/swh/web/tests/common/test_service.py @@ -1,950 +1,971 @@ # Copyright (C) 2015-2020 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 itertools import pytest import random from collections import defaultdict from hypothesis import given from swh.model.hashutil import hash_to_bytes, hash_to_hex from swh.model.from_disk import DentryPerms from swh.model.identifiers import CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT from swh.model.model import Directory, DirectoryEntry, Origin, Revision from swh.web.common import service from swh.web.common.exc import BadInputExc, NotFoundExc from swh.web.tests.data import random_sha1, random_content from swh.web.tests.strategies import ( content, unknown_content, contents, unknown_contents, contents_with_ctags, origin, new_origin, visit_dates, directory, unknown_directory, release, unknown_release, revision, unknown_revision, revisions, ancestor_revisions, non_ancestor_revisions, invalid_sha1, sha256, revision_with_submodules, empty_directory, new_revision, snapshot, unknown_snapshot, ) from swh.web.tests.conftest import ctags_json_missing, fossology_missing @given(contents()) 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 service.lookup_multiple_hashes(input_data) == expected_output @given(contents(), unknown_contents()) 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 service.lookup_multiple_hashes(input_data) == expected_output def test_lookup_hash_does_not_exist(): unknown_content_ = random_content() actual_lookup = service.lookup_hash("sha1_git:%s" % unknown_content_["sha1_git"]) assert actual_lookup == {"found": None, "algo": "sha1_git"} @given(content()) def test_lookup_hash_exist(archive_data, content): actual_lookup = service.lookup_hash("sha1:%s" % content["sha1"]) content_metadata = archive_data.content_get_metadata(content["sha1"]) assert {"found": content_metadata, "algo": "sha1"} == actual_lookup def test_search_hash_does_not_exist(): unknown_content_ = random_content() actual_lookup = service.search_hash("sha1_git:%s" % unknown_content_["sha1_git"]) assert {"found": False} == actual_lookup @given(content()) def test_search_hash_exist(content): actual_lookup = service.search_hash("sha1:%s" % content["sha1"]) assert {"found": True} == actual_lookup @pytest.mark.skipif( ctags_json_missing, reason="requires ctags with json output support" ) @given(contents_with_ctags()) 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(service.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( service.lookup_content_ctags("sha1:%s" % unknown_content_["sha1"]) ) assert actual_ctags == [] @given(content()) def test_lookup_content_filetype(indexer_data, content): indexer_data.content_add_mimetype(content["sha1"]) actual_filetype = service.lookup_content_filetype(content["sha1"]) expected_filetype = indexer_data.content_get_mimetype(content["sha1"]) assert actual_filetype == expected_filetype @pytest.mark.skip # Language indexer is disabled. @given(content()) def test_lookup_content_language(indexer_data, content): indexer_data.content_add_language(content["sha1"]) actual_language = service.lookup_content_language(content["sha1"]) expected_language = indexer_data.content_get_language(content["sha1"]) assert actual_language == expected_language @given(contents_with_ctags()) 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( service.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( service.lookup_expression("barfoo", last_sha1=None, per_page=10) ) assert actual_ctags == expected_ctags @pytest.mark.skipif(fossology_missing, reason="requires fossology-nomossa installed") @given(content()) def test_lookup_content_license(indexer_data, content): indexer_data.content_add_license(content["sha1"]) actual_license = service.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 = service.stat_counters() assert actual_stats == archive_data.stat_counters() @given(new_origin(), visit_dates()) def test_lookup_origin_visits(archive_data, new_origin, visit_dates): archive_data.origin_add_one(new_origin) for ts in visit_dates: archive_data.origin_visit_add(new_origin.url, ts, type="git") actual_origin_visits = list( service.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_one(new_origin) visits = [] for ts in visit_dates: visits.append(archive_data.origin_visit_add(new_origin.url, ts, type="git")) visit = random.choice(visits).visit actual_origin_visit = service.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()) def test_lookup_origin(archive_data, new_origin): archive_data.origin_add_one(new_origin) actual_origin = service.lookup_origin({"url": new_origin.url}) expected_origin = archive_data.origin_get({"url": new_origin.url}) assert actual_origin == expected_origin @given(invalid_sha1()) def test_lookup_release_ko_id_checksum_not_a_sha1(invalid_sha1): with pytest.raises(BadInputExc) as e: service.lookup_release(invalid_sha1) assert e.match("Invalid checksum") @given(sha256()) def test_lookup_release_ko_id_checksum_too_long(sha256): with pytest.raises(BadInputExc) as e: service.lookup_release(sha256) assert e.match("Only sha1_git is supported.") @given(directory()) def test_lookup_directory_with_path_not_found(directory): path = "some/invalid/path/here" with pytest.raises(NotFoundExc) as e: service.lookup_directory_with_path(directory, path) assert e.match("Directory entry with path %s from %s not found" % (path, directory)) @given(directory()) 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 = service.lookup_directory_with_path(directory, path) assert actual_result == directory_entry @given(release()) def test_lookup_release(archive_data, release): actual_release = service.lookup_release(release) assert actual_release == archive_data.release_get(release) @given(revision(), invalid_sha1(), sha256()) 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: service.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: service.lookup_revision_with_context(sha1_git_root, sha1_git) assert e.match("Only sha1_git is supported") @given(revision(), unknown_revision()) 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: service.lookup_revision_with_context(sha1_git_root, sha1_git) assert e.match("Revision %s not found" % sha1_git) @given(revision(), unknown_revision()) 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: service.lookup_revision_with_context(sha1_git_root, sha1_git) assert e.match("Revision root %s not found" % sha1_git_root) @given(ancestor_revisions()) 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 = service.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 @given(non_ancestor_revisions()) 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: service.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: service.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: service.lookup_directory_with_revision(new_revision_id, dir_path) assert e.match("Content not found for revision %s" % new_revision_id) @given(revision()) def test_lookup_directory_with_revision_ko_path_to_nowhere(revision): invalid_path = "path/to/something/unknown" with pytest.raises(NotFoundExc) as e: service.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") @given(revision_with_submodules()) 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 = service.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 @given(revision()) def test_lookup_directory_with_revision_without_path(archive_data, revision): actual_directory_entries = service.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 @given(revision()) 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 = service.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 @given(revision()) 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(expected_dir_entry["checksums"]["sha1"]) actual_dir_entry = service.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"] @given(revision()) def test_lookup_revision(archive_data, revision): actual_revision = service.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 = service.lookup_revision(hash_to_hex(new_revision["id"])) assert revision["message"] is None assert revision["message_decoding_failed"] is True @given(new_revision()) def test_lookup_revision_msg_ok(archive_data, new_revision): archive_data.revision_add([new_revision]) revision_message = service.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: service.lookup_revision_message(unknown_revision_) assert e.match("Revision with sha1_git %s not found." % unknown_revision_) @given(revisions()) def test_lookup_revision_multiple(archive_data, revisions): actual_revisions = list(service.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(service.lookup_revision_multiple(unknown_revisions_)) assert actual_revisions == [None] * len(unknown_revisions_) @given(revision()) def test_lookup_revision_log(archive_data, revision): actual_revision_log = list(service.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 @given(origin()) 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( service.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 @given(origin()) def test_lookup_revision_log_by_notfound(origin): with pytest.raises(NotFoundExc): service.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: service.lookup_content_raw("sha1:" + unknown_content_["sha1"]) assert e.match( "Content with %s checksum equals to %s not found!" % ("sha1", unknown_content_["sha1"]) ) @given(content()) def test_lookup_content_raw(archive_data, content): actual_content = service.lookup_content_raw("sha256:%s" % content["sha256"]) expected_content = archive_data.content_get(content["sha1"]) assert actual_content == expected_content def test_lookup_content_not_found(): unknown_content_ = random_content() with pytest.raises(NotFoundExc) as e: service.lookup_content("sha1:%s" % unknown_content_["sha1"]) assert e.match( "Content with %s checksum equals to %s not found!" % ("sha1", unknown_content_["sha1"]) ) @given(content()) def test_lookup_content_with_sha1(archive_data, content): actual_content = service.lookup_content("sha1:%s" % content["sha1"]) expected_content = archive_data.content_get_metadata(content["sha1"]) assert actual_content == expected_content @given(content()) def test_lookup_content_with_sha256(archive_data, content): actual_content = service.lookup_content("sha256:%s" % content["sha256"]) expected_content = archive_data.content_get_metadata(content["sha1"]) assert actual_content == expected_content def test_lookup_directory_bad_checksum(): with pytest.raises(BadInputExc): service.lookup_directory("directory_id") def test_lookup_directory_not_found(): unknown_directory_ = random_sha1() with pytest.raises(NotFoundExc) as e: service.lookup_directory(unknown_directory_) assert e.match("Directory with sha1_git %s not found" % unknown_directory_) @given(directory()) def test_lookup_directory(archive_data, directory): actual_directory_ls = list(service.lookup_directory(directory)) expected_directory_ls = archive_data.directory_ls(directory) assert actual_directory_ls == expected_directory_ls @given(empty_directory()) def test_lookup_directory_empty(empty_directory): actual_directory_ls = list(service.lookup_directory(empty_directory)) assert actual_directory_ls == [] @given(origin()) def test_lookup_revision_by_nothing_found(origin): with pytest.raises(NotFoundExc): service.lookup_revision_by(origin["url"], "invalid-branch-name") @given(origin()) def test_lookup_revision_by(archive_data, origin): branches = _get_origin_branches(archive_data, origin) branch_name = random.choice(list(branches.keys())) actual_revision = service.lookup_revision_by(origin["url"], branch_name) expected_revision = archive_data.revision_get(branches[branch_name]["target"]) assert actual_revision == expected_revision @given(origin(), revision()) def test_lookup_revision_with_context_by_ko(origin, revision): with pytest.raises(NotFoundExc): service.lookup_revision_with_context_by( origin["url"], "invalid-branch-name", None, revision ) @given(origin()) 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 = service.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): service.lookup_revision_through({"something-unknown": 10}) @given(origin()) 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 service.lookup_revision_through( { "origin_url": origin["url"], "branch_name": branch_name, "ts": None, "sha1_git": rev, } ) == service.lookup_revision_with_context_by(origin["url"], branch_name, None, rev) @given(origin()) 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 service.lookup_revision_through( {"origin_url": origin["url"], "branch_name": branch_name, "ts": None,} ) == service.lookup_revision_by(origin["url"], branch_name, None) @given(ancestor_revisions()) def test_lookup_revision_through_with_context(ancestor_revisions): sha1_git = ancestor_revisions["sha1_git"] sha1_git_root = ancestor_revisions["sha1_git_root"] assert service.lookup_revision_through( {"sha1_git_root": sha1_git_root, "sha1_git": sha1_git,} ) == service.lookup_revision_with_context(sha1_git_root, sha1_git) @given(revision()) def test_lookup_revision_through_with_revision(revision): assert service.lookup_revision_through( {"sha1_git": revision} ) == service.lookup_revision(revision) @given(revision()) def test_lookup_directory_through_revision_ko_not_found(revision): with pytest.raises(NotFoundExc): service.lookup_directory_through_revision( {"sha1_git": revision}, "some/invalid/path" ) @given(revision()) 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 service.lookup_directory_through_revision( {"sha1_git": revision}, dir_entry["name"] ) == (revision, service.lookup_directory_with_revision(revision, dir_entry["name"])) @given(revision()) 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 service.lookup_directory_through_revision( {"sha1_git": revision}, dir_entry["name"], with_data=True ) == ( revision, service.lookup_directory_with_revision( revision, dir_entry["name"], with_data=True ), ) @given(content(), directory(), release(), revision(), snapshot()) def test_lookup_known_objects( archive_data, content, directory, release, revision, snapshot ): expected = archive_data.content_find(content) assert service.lookup_object(CONTENT, content["sha1_git"]) == expected expected = archive_data.directory_get(directory) assert service.lookup_object(DIRECTORY, directory) == expected expected = archive_data.release_get(release) assert service.lookup_object(RELEASE, release) == expected expected = archive_data.revision_get(revision) assert service.lookup_object(REVISION, revision) == expected expected = archive_data.snapshot_get(snapshot) assert service.lookup_object(SNAPSHOT, snapshot) == expected @given( unknown_content(), unknown_directory(), unknown_release(), unknown_revision(), unknown_snapshot(), ) def test_lookup_unknown_objects( unknown_content, unknown_directory, unknown_release, unknown_revision, unknown_snapshot, ): with pytest.raises(NotFoundExc) as e: service.lookup_object(CONTENT, unknown_content["sha1_git"]) assert e.match(r"Content.*not found") with pytest.raises(NotFoundExc) as e: service.lookup_object(DIRECTORY, unknown_directory) assert e.match(r"Directory.*not found") with pytest.raises(NotFoundExc) as e: service.lookup_object(RELEASE, unknown_release) assert e.match(r"Release.*not found") with pytest.raises(NotFoundExc) as e: service.lookup_object(REVISION, unknown_revision) assert e.match(r"Revision.*not found") with pytest.raises(NotFoundExc) as e: service.lookup_object(SNAPSHOT, unknown_snapshot) assert e.match(r"Snapshot.*not found") @given(invalid_sha1()) def test_lookup_invalid_objects(invalid_sha1): with pytest.raises(BadInputExc) as e: service.lookup_object("foo", invalid_sha1) assert e.match("Invalid swh object type") with pytest.raises(BadInputExc) as e: service.lookup_object(CONTENT, invalid_sha1) assert e.match("Invalid hash") with pytest.raises(BadInputExc) as e: service.lookup_object(DIRECTORY, invalid_sha1) assert e.match("Invalid checksum") with pytest.raises(BadInputExc) as e: service.lookup_object(RELEASE, invalid_sha1) assert e.match("Invalid checksum") with pytest.raises(BadInputExc) as e: service.lookup_object(REVISION, invalid_sha1) assert e.match("Invalid checksum") with pytest.raises(BadInputExc) as e: service.lookup_object(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_pids = { CONTENT: [hash_to_bytes(missing_cnt)], DIRECTORY: [hash_to_bytes(missing_dir)], REVISION: [hash_to_bytes(missing_rev)], RELEASE: [hash_to_bytes(missing_rel)], SNAPSHOT: [hash_to_bytes(missing_snp)], } actual_result = service.lookup_missing_hashes(grouped_pids) assert actual_result == { missing_cnt, missing_dir, missing_rev, missing_rel, missing_snp, } @given(content(), directory()) def test_lookup_missing_hashes_some_present(archive_data, content, directory): missing_rev = random_sha1() missing_rel = random_sha1() missing_snp = random_sha1() grouped_pids = { CONTENT: [hash_to_bytes(content["sha1_git"])], DIRECTORY: [hash_to_bytes(directory)], REVISION: [hash_to_bytes(missing_rev)], RELEASE: [hash_to_bytes(missing_rel)], SNAPSHOT: [hash_to_bytes(missing_snp)], } actual_result = service.lookup_missing_hashes(grouped_pids) assert actual_result == {missing_rev, missing_rel, missing_snp} @given(origin()) def test_lookup_origin_extra_trailing_slash(origin): origin_info = service.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_one(deb_origin) origin_info = service.lookup_origin({"url": deb_origin.url[:-1]}) assert origin_info["url"] == deb_origin.url + + +@given(snapshot()) +def test_lookup_snapshot_branch_name_from_tip_revision(archive_data, snapshot_id): + snapshot = archive_data.snapshot_get(snapshot_id) + branches = [ + {"name": k, "revision": v["target"]} + for k, v in snapshot["branches"].items() + if v["target_type"] == "revision" + ] + branch_info = random.choice(branches) + possible_results = [ + b["name"] for b in branches if b["revision"] == branch_info["revision"] + ] + + assert ( + service.lookup_snapshot_branch_name_from_tip_revision( + snapshot_id, branch_info["revision"] + ) + in possible_results + )