diff --git a/swh/web/api/views/identifiers.py b/swh/web/api/views/identifiers.py index 4281f9a2..ac8e4a04 100644 --- a/swh/web/api/views/identifiers.py +++ b/swh/web/api/views/identifiers.py @@ -1,120 +1,120 @@ -# Copyright (C) 2018-2020 The Software Heritage developers +# Copyright (C) 2018-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from swh.model.hashutil import hash_to_bytes, hash_to_hex from swh.web.api.apidoc import api_doc, format_docstring from swh.web.api.apiurls import api_route from swh.web.common import archive from swh.web.common.exc import LargePayloadExc from swh.web.common.identifiers import get_swhid, group_swhids, resolve_swhid @api_route(r"/resolve/(?P.*)/", "api-1-resolve-swhid") @api_doc("/resolve/") @format_docstring() def api_resolve_swhid(request, swhid): """ .. http:get:: /api/1/resolve/(swhid)/ Resolve a SoftWare Heritage persistent IDentifier (SWHID) Try to resolve a provided `SoftWare Heritage persistent IDentifier `_ into an url for browsing the pointed archive object. If the provided identifier is valid, the existence of the object in the archive will also be checked. :param string swhid: a SoftWare Heritage persistent IDentifier :>json string browse_url: the url for browsing the pointed object :>json object metadata: object holding optional parts of the SWHID :>json string namespace: the SWHID namespace :>json string object_id: the hash identifier of the pointed object :>json string object_type: the type of the pointed object :>json number scheme_version: the scheme version of the SWHID {common_headers} :statuscode 200: no error :statuscode 400: an invalid SWHID has been provided :statuscode 404: the pointed object does not exist in the archive **Example:** .. parsed-literal:: :swh_web_api:`resolve/swh:1:rev:96db9023b881d7cd9f379b0c154650d6c108e9a3;origin=https://github.com/openssl/openssl/` """ # try to resolve the provided swhid swhid_resolved = resolve_swhid(swhid) # id is well-formed, now check that the pointed # object is present in the archive, NotFoundExc # will be raised otherwise swhid_parsed = swhid_resolved["swhid_parsed"] - object_type = swhid_parsed.object_type.name.lower() + object_type = swhid_parsed.object_type object_id = hash_to_hex(swhid_parsed.object_id) - archive.lookup_object(object_type, object_id) + archive.lookup_object(swhid_parsed.object_type, object_id) # id is well-formed and the pointed object exists return { "namespace": swhid_parsed.namespace, "scheme_version": swhid_parsed.scheme_version, - "object_type": object_type, + "object_type": object_type.name.lower(), "object_id": object_id, "metadata": swhid_parsed.qualifiers(), "browse_url": request.build_absolute_uri(swhid_resolved["browse_url"]), } @api_route(r"/known/", "api-1-known", methods=["POST"]) @api_doc("/known/") @format_docstring() def api_swhid_known(request): """ .. http:post:: /api/1/known/ Check if a list of objects are present in the Software Heritage archive. The objects to check existence must be provided using `SoftWare Heritage persistent IDentifiers `_. :json object : an object whose keys are input SWHIDs and values objects with the following keys: * **known (bool)**: whether the object was found {common_headers} :statuscode 200: no error :statuscode 400: an invalid SWHID was provided :statuscode 413: the input array of SWHIDs is too large """ limit = 1000 if len(request.data) > limit: raise LargePayloadExc( "The maximum number of SWHIDs this endpoint can receive is %s" % limit ) swhids = [get_swhid(swhid) for swhid in request.data] response = {str(swhid): {"known": False} for swhid in swhids} # group swhids by their type swhids_by_type = group_swhids(swhids) # search for hashes not present in the storage missing_hashes = set( map(hash_to_bytes, archive.lookup_missing_hashes(swhids_by_type)) ) for swhid in swhids: if swhid.object_id not in missing_hashes: response[str(swhid)]["known"] = True return response diff --git a/swh/web/browse/snapshot_context.py b/swh/web/browse/snapshot_context.py index 39577e1b..76d91c5b 100644 --- a/swh/web/browse/snapshot_context.py +++ b/swh/web/browse/snapshot_context.py @@ -1,1471 +1,1477 @@ # Copyright (C) 2018-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information # Utility module for browsing the archive in a snapshot context. from collections import defaultdict from typing import Any, Dict, List, Optional, Tuple from django.core.cache import cache from django.shortcuts import render from django.utils.html import escape from swh.model.hashutil import hash_to_bytes -from swh.model.identifiers import ( - CONTENT, - DIRECTORY, - RELEASE, - REVISION, - SNAPSHOT, - CoreSWHID, - ObjectType, -) +from swh.model.identifiers import CoreSWHID, ObjectType from swh.model.model import Snapshot from swh.web.browse.utils import ( content_display_max_size, format_log_entries, gen_content_link, gen_release_link, gen_revision_link, gen_revision_log_link, gen_revision_url, gen_snapshot_link, get_directory_entries, get_readme_to_display, prepare_content_for_display, request_content, ) from swh.web.common import archive, highlightjs from swh.web.common.exc import BadInputExc, NotFoundExc, http_status_code_message from swh.web.common.identifiers import get_swhids_info from swh.web.common.origin_visits import get_origin_visit from swh.web.common.typing import ( ContentMetadata, DirectoryMetadata, OriginInfo, SnapshotBranchInfo, SnapshotContext, SnapshotReleaseInfo, SWHObjectInfo, ) from swh.web.common.utils import ( format_utc_iso_date, gen_path_info, reverse, swh_object_icons, ) from swh.web.config import get_config _empty_snapshot_id = Snapshot(branches={}).id.hex() def _get_branch(branches, branch_name, snapshot_id): """ Utility function to get a specific branch from a branches list. Its purpose is to get the default HEAD branch as some software origin (e.g those with svn type) does not have it. In that latter case, check if there is a master branch instead and returns it. """ filtered_branches = [b for b in branches if b["name"] == branch_name] if filtered_branches: return filtered_branches[0] elif branch_name == "HEAD": filtered_branches = [b for b in branches if b["name"].endswith("master")] if filtered_branches: return filtered_branches[0] elif branches: return branches[0] else: # case where a large branches list has been truncated snp = archive.lookup_snapshot( snapshot_id, branches_from=branch_name, branches_count=1, target_types=["revision", "alias"], # pull request branches must be browsable even if they are hidden # by default in branches list branch_name_exclude_prefix=None, ) snp_branch, _, _ = process_snapshot_branches(snp) if snp_branch and snp_branch[0]["name"] == branch_name: branches.append(snp_branch[0]) return snp_branch[0] def _get_release(releases, release_name, snapshot_id): """ Utility function to get a specific release from a releases list. Returns None if the release can not be found in the list. """ filtered_releases = [r for r in releases if r["name"] == release_name] if filtered_releases: return filtered_releases[0] else: # case where a large branches list has been truncated try: # git origins have specific branches for releases snp = archive.lookup_snapshot( snapshot_id, branches_from=f"refs/tags/{release_name}", branches_count=1, target_types=["release"], ) except NotFoundExc: snp = archive.lookup_snapshot( snapshot_id, branches_from=release_name, branches_count=1, target_types=["release"], ) _, snp_release, _ = process_snapshot_branches(snp) if snp_release and snp_release[0]["name"] == release_name: releases.append(snp_release[0]) return snp_release[0] def _branch_not_found( branch_type, branch, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id ): """ Utility function to raise an exception when a specified branch/release can not be found. """ if branch_type == "branch": branch_type = "Branch" branch_type_plural = "branches" target_type = "revision" else: branch_type = "Release" branch_type_plural = "releases" target_type = "release" if snapshot_id and snapshot_sizes[target_type] == 0: msg = "Snapshot with id %s has an empty list" " of %s!" % ( snapshot_id, branch_type_plural, ) elif snapshot_id: msg = "%s %s for snapshot with id %s" " not found!" % ( branch_type, branch, snapshot_id, ) elif visit_id and snapshot_sizes[target_type] == 0: msg = ( "Origin with url %s" " for visit with id %s has an empty list" " of %s!" % (origin_info["url"], visit_id, branch_type_plural) ) elif visit_id: msg = ( "%s %s associated to visit with" " id %s for origin with url %s" " not found!" % (branch_type, branch, visit_id, origin_info["url"]) ) elif snapshot_sizes[target_type] == 0: msg = ( "Origin with url %s" " for visit with timestamp %s has an empty list" " of %s!" % (origin_info["url"], timestamp, branch_type_plural) ) else: msg = ( "%s %s associated to visit with" " timestamp %s for origin with " "url %s not found!" % (branch_type, branch, timestamp, origin_info["url"]) ) raise NotFoundExc(escape(msg)) def process_snapshot_branches( snapshot: Dict[str, Any] ) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo], Dict[str, Any]]: """ Process a dictionary describing snapshot branches: extract those targeting revisions and releases, put them in two different lists, then sort those lists in lexicographical order of the branches' names. Args: snapshot: A dict describing a snapshot as returned for instance by :func:`swh.web.common.archive.lookup_snapshot` Returns: A tuple whose first member is the sorted list of branches targeting revisions, second member the sorted list of branches targeting releases and third member a dict mapping resolved branch aliases to their real target. """ snapshot_branches = snapshot["branches"] branches: Dict[str, SnapshotBranchInfo] = {} branch_aliases: Dict[str, str] = {} releases: Dict[str, SnapshotReleaseInfo] = {} revision_to_branch = defaultdict(set) revision_to_release = defaultdict(set) release_to_branch = defaultdict(set) for branch_name, target in snapshot_branches.items(): if not target: # FIXME: display branches with an unknown target anyway continue target_id = target["target"] target_type = target["target_type"] if target_type == "revision": branches[branch_name] = SnapshotBranchInfo( name=branch_name, alias=False, revision=target_id, date=None, directory=None, message=None, url=None, ) revision_to_branch[target_id].add(branch_name) elif target_type == "release": release_to_branch[target_id].add(branch_name) elif target_type == "alias": branch_aliases[branch_name] = target_id # FIXME: handle pointers to other object types def _add_release_info(branch, release, alias=False): releases[branch] = SnapshotReleaseInfo( name=release["name"], alias=alias, branch_name=branch, date=format_utc_iso_date(release["date"]), directory=None, id=release["id"], message=release["message"], target_type=release["target_type"], target=release["target"], url=None, ) def _add_branch_info(branch, revision, alias=False): branches[branch] = SnapshotBranchInfo( name=branch, alias=alias, revision=revision["id"], directory=revision["directory"], date=format_utc_iso_date(revision["date"]), message=revision["message"], url=None, ) releases_info = archive.lookup_release_multiple(release_to_branch.keys()) for release in releases_info: if release is None: continue branches_to_update = release_to_branch[release["id"]] for branch in branches_to_update: _add_release_info(branch, release) if release["target_type"] == "revision": revision_to_release[release["target"]].update(branches_to_update) revisions = archive.lookup_revision_multiple( set(revision_to_branch.keys()) | set(revision_to_release.keys()) ) for revision in revisions: if not revision: continue for branch in revision_to_branch[revision["id"]]: _add_branch_info(branch, revision) for release_id in revision_to_release[revision["id"]]: releases[release_id]["directory"] = revision["directory"] resolved_aliases = {} for branch_alias, branch_target in branch_aliases.items(): resolved_alias = archive.lookup_snapshot_alias(snapshot["id"], branch_alias) resolved_aliases[branch_alias] = resolved_alias if resolved_alias is None: continue target_type = resolved_alias["target_type"] target = resolved_alias["target"] if target_type == "revision": revision = archive.lookup_revision(target) _add_branch_info(branch_alias, revision, alias=True) elif target_type == "release": release = archive.lookup_release(target) _add_release_info(branch_alias, release, alias=True) if branch_alias in branches: branches[branch_alias]["name"] = branch_alias ret_branches = list(sorted(branches.values(), key=lambda b: b["name"])) ret_releases = list(sorted(releases.values(), key=lambda b: b["name"])) return ret_branches, ret_releases, resolved_aliases def get_snapshot_content( snapshot_id: str, ) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo], Dict[str, Any]]: """Returns the lists of branches and releases associated to a swh snapshot. That list is put in cache in order to speedup the navigation in the swh-web/browse ui. .. warning:: At most 1000 branches contained in the snapshot will be returned for performance reasons. Args: snapshot_id: hexadecimal representation of the snapshot identifier Returns: A tuple with three members. The first one is a list of dict describing the snapshot branches. The second one is a list of dict describing the snapshot releases. The third one is a dict mapping resolved branch aliases to their real target. Raises: NotFoundExc if the snapshot does not exist """ cache_entry_id = "swh_snapshot_%s" % snapshot_id cache_entry = cache.get(cache_entry_id) if cache_entry: return ( cache_entry["branches"], cache_entry["releases"], cache_entry.get("aliases", {}), ) branches: List[SnapshotBranchInfo] = [] releases: List[SnapshotReleaseInfo] = [] aliases: Dict[str, Any] = {} snapshot_content_max_size = get_config()["snapshot_content_max_size"] if snapshot_id: snapshot = archive.lookup_snapshot( snapshot_id, branches_count=snapshot_content_max_size ) branches, releases, aliases = process_snapshot_branches(snapshot) cache.set( cache_entry_id, {"branches": branches, "releases": releases, "aliases": aliases} ) return branches, releases, aliases def get_origin_visit_snapshot( origin_info: OriginInfo, visit_ts: Optional[str] = None, visit_id: Optional[int] = None, snapshot_id: Optional[str] = None, ) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo], Dict[str, Any]]: """Returns the lists of branches and releases associated to an origin for a given visit. The visit is expressed by either: * a snapshot identifier * a timestamp, if no visit with that exact timestamp is found, the closest one from the provided timestamp will be used. If no visit parameter is provided, it returns the list of branches found for the latest visit. That list is put in cache in order to speedup the navigation in the swh-web/browse ui. .. warning:: At most 1000 branches contained in the snapshot will be returned for performance reasons. Args: origin_info: a dict filled with origin information visit_ts: an ISO 8601 datetime string to parse visit_id: visit id for disambiguation in case several visits have the same timestamp snapshot_id: if provided, visit associated to the snapshot will be processed Returns: A tuple with three members. The first one is a list of dict describing the origin branches for the given visit. The second one is a list of dict describing the origin releases for the given visit. The third one is a dict mapping resolved branch aliases to their real target. Raises: NotFoundExc if the origin or its visit are not found """ visit_info = get_origin_visit(origin_info, visit_ts, visit_id, snapshot_id) return get_snapshot_content(visit_info["snapshot"]) def get_snapshot_context( snapshot_id: Optional[str] = None, origin_url: Optional[str] = None, timestamp: Optional[str] = None, visit_id: Optional[int] = None, branch_name: Optional[str] = None, release_name: Optional[str] = None, revision_id: Optional[str] = None, path: Optional[str] = None, browse_context: str = "directory", ) -> SnapshotContext: """ Utility function to compute relevant information when navigating the archive in a snapshot context. The snapshot is either referenced by its id or it will be retrieved from an origin visit. Args: snapshot_id: hexadecimal representation of a snapshot identifier origin_url: an origin_url timestamp: a datetime string for retrieving the closest visit of the origin visit_id: optional visit id for disambiguation in case of several visits with the same timestamp branch_name: optional branch name set when browsing the snapshot in that scope (will default to "HEAD" if not provided) release_name: optional release name set when browsing the snapshot in that scope revision_id: optional revision identifier set when browsing the snapshot in that scope path: optional path of the object currently browsed in the snapshot browse_context: indicates which type of object is currently browsed Returns: A dict filled with snapshot context information. Raises: swh.web.common.exc.NotFoundExc: if no snapshot is found for the visit of an origin. """ assert origin_url is not None or snapshot_id is not None origin_info = None visit_info = None url_args = {} query_params: Dict[str, Any] = {} origin_visits_url = None if origin_url: if visit_id is not None: query_params["visit_id"] = visit_id elif snapshot_id is not None: query_params["snapshot"] = snapshot_id origin_info = archive.lookup_origin({"url": origin_url}) visit_info = get_origin_visit(origin_info, timestamp, visit_id, snapshot_id) formatted_date = format_utc_iso_date(visit_info["date"]) visit_info["formatted_date"] = formatted_date snapshot_id = visit_info["snapshot"] if not snapshot_id: raise NotFoundExc( "No snapshot associated to the visit of origin " "%s on %s" % (escape(origin_url), formatted_date) ) # provided timestamp is not necessarily equals to the one # of the retrieved visit, so get the exact one in order # to use it in the urls generated below if timestamp: timestamp = visit_info["date"] branches, releases, aliases = get_origin_visit_snapshot( origin_info, timestamp, visit_id, snapshot_id ) query_params["origin_url"] = origin_info["url"] origin_visits_url = reverse( "browse-origin-visits", query_params={"origin_url": origin_info["url"]} ) if timestamp is not None: query_params["timestamp"] = format_utc_iso_date( timestamp, "%Y-%m-%dT%H:%M:%SZ" ) visit_url = reverse("browse-origin-directory", query_params=query_params) visit_info["url"] = visit_url branches_url = reverse("browse-origin-branches", query_params=query_params) releases_url = reverse("browse-origin-releases", query_params=query_params) else: assert snapshot_id is not None branches, releases, aliases = get_snapshot_content(snapshot_id) url_args = {"snapshot_id": snapshot_id} branches_url = reverse("browse-snapshot-branches", url_args=url_args) releases_url = reverse("browse-snapshot-releases", url_args=url_args) releases = list(reversed(releases)) snapshot_sizes_cache_id = f"swh_snapshot_{snapshot_id}_sizes" snapshot_sizes = cache.get(snapshot_sizes_cache_id) if snapshot_sizes is None: snapshot_sizes = archive.lookup_snapshot_sizes(snapshot_id) cache.set(snapshot_sizes_cache_id, snapshot_sizes) is_empty = (snapshot_sizes["release"] + snapshot_sizes["revision"]) == 0 swh_snp_id = str( CoreSWHID(object_type=ObjectType.SNAPSHOT, object_id=hash_to_bytes(snapshot_id)) ) if visit_info: timestamp = format_utc_iso_date(visit_info["date"]) if origin_info: browse_view_name = f"browse-origin-{browse_context}" else: browse_view_name = f"browse-snapshot-{browse_context}" release_id = None root_directory = None snapshot_total_size = snapshot_sizes["release"] + snapshot_sizes["revision"] if path is not None: query_params["path"] = path if snapshot_total_size and revision_id is not None: revision = archive.lookup_revision(revision_id) root_directory = revision["directory"] branches.append( SnapshotBranchInfo( name=revision_id, alias=False, revision=revision_id, directory=root_directory, date=revision["date"], message=revision["message"], url=None, ) ) query_params["revision"] = revision_id elif snapshot_total_size and release_name: release = _get_release(releases, release_name, snapshot_id) if release is None: _branch_not_found( "release", release_name, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id, ) else: root_directory = release["directory"] revision_id = release["target"] release_id = release["id"] query_params["release"] = release_name elif snapshot_total_size: if branch_name: query_params["branch"] = branch_name branch = _get_branch(branches, branch_name or "HEAD", snapshot_id) if branch is None: _branch_not_found( "branch", branch_name, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id, ) else: branch_name = branch["name"] revision_id = branch["revision"] root_directory = branch["directory"] for b in branches: branch_query_params = dict(query_params) branch_query_params.pop("release", None) if b["name"] != b["revision"]: branch_query_params.pop("revision", None) branch_query_params["branch"] = b["name"] b["url"] = reverse( browse_view_name, url_args=url_args, query_params=branch_query_params ) for r in releases: release_query_params = dict(query_params) release_query_params.pop("branch", None) release_query_params.pop("revision", None) release_query_params["release"] = r["name"] r["url"] = reverse( browse_view_name, url_args=url_args, query_params=release_query_params, ) revision_info = None if revision_id: try: revision_info = archive.lookup_revision(revision_id) except NotFoundExc: pass else: revision_info["date"] = format_utc_iso_date(revision_info["date"]) revision_info["committer_date"] = format_utc_iso_date( revision_info["committer_date"] ) if revision_info["message"]: message_lines = revision_info["message"].split("\n") revision_info["message_header"] = message_lines[0] else: revision_info["message_header"] = "" snapshot_context = SnapshotContext( branch=branch_name, branch_alias=branch_name in aliases, branches=branches, branches_url=branches_url, is_empty=is_empty, origin_info=origin_info, origin_visits_url=origin_visits_url, release=release_name, release_alias=release_name in aliases, release_id=release_id, query_params=query_params, releases=releases, releases_url=releases_url, revision_id=revision_id, revision_info=revision_info, root_directory=root_directory, snapshot_id=snapshot_id, snapshot_sizes=snapshot_sizes, snapshot_swhid=swh_snp_id, url_args=url_args, visit_info=visit_info, ) if revision_info: revision_info["revision_url"] = gen_revision_url(revision_id, snapshot_context) return snapshot_context def _build_breadcrumbs(snapshot_context: SnapshotContext, path: str): origin_info = snapshot_context["origin_info"] url_args = snapshot_context["url_args"] query_params = dict(snapshot_context["query_params"]) root_directory = snapshot_context["root_directory"] path_info = gen_path_info(path) if origin_info: browse_view_name = "browse-origin-directory" else: browse_view_name = "browse-snapshot-directory" breadcrumbs = [] if root_directory: query_params.pop("path", None) breadcrumbs.append( { "name": root_directory[:7], "url": reverse( browse_view_name, url_args=url_args, query_params=query_params ), } ) for pi in path_info: query_params["path"] = pi["path"] breadcrumbs.append( { "name": pi["name"], "url": reverse( browse_view_name, url_args=url_args, query_params=query_params ), } ) return breadcrumbs def _check_origin_url(snapshot_id, origin_url): if snapshot_id is None and origin_url is None: raise BadInputExc("An origin URL must be provided as query parameter.") def browse_snapshot_directory( request, snapshot_id=None, origin_url=None, timestamp=None, path=None ): """ Django view implementation for browsing a directory in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), path=path, browse_context="directory", branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), ) root_directory = snapshot_context["root_directory"] sha1_git = root_directory error_info = { "status_code": 200, "description": None, } if root_directory and path: try: dir_info = archive.lookup_directory_with_path(root_directory, path) sha1_git = dir_info["target"] except NotFoundExc as e: sha1_git = None error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" dirs = [] files = [] if sha1_git: dirs, files = get_directory_entries(sha1_git) origin_info = snapshot_context["origin_info"] visit_info = snapshot_context["visit_info"] url_args = snapshot_context["url_args"] query_params = dict(snapshot_context["query_params"]) revision_id = snapshot_context["revision_id"] snapshot_id = snapshot_context["snapshot_id"] if origin_info: browse_view_name = "browse-origin-directory" else: browse_view_name = "browse-snapshot-directory" breadcrumbs = _build_breadcrumbs(snapshot_context, path) path = "" if path is None else (path + "/") for d in dirs: if d["type"] == "rev": d["url"] = reverse("browse-revision", url_args={"sha1_git": d["target"]}) else: query_params["path"] = path + d["name"] d["url"] = reverse( browse_view_name, url_args=url_args, query_params=query_params ) sum_file_sizes = 0 readmes = {} if origin_info: browse_view_name = "browse-origin-content" else: browse_view_name = "browse-snapshot-content" for f in files: query_params["path"] = path + f["name"] f["url"] = reverse( browse_view_name, url_args=url_args, query_params=query_params ) if f["length"] is not None: sum_file_sizes += f["length"] if f["name"].lower().startswith("readme"): readmes[f["name"]] = f["checksums"]["sha1"] readme_name, readme_url, readme_html = get_readme_to_display(readmes) if origin_info: browse_view_name = "browse-origin-log" else: browse_view_name = "browse-snapshot-log" history_url = None if snapshot_id != _empty_snapshot_id: query_params.pop("path", None) history_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) nb_files = None nb_dirs = None dir_path = None if root_directory: nb_files = len(files) nb_dirs = len(dirs) dir_path = "/" + path swh_objects = [] vault_cooking = {} revision_found = True if sha1_git is None and revision_id is not None: try: archive.lookup_revision(revision_id) except NotFoundExc: revision_found = False if sha1_git is not None: - swh_objects.append(SWHObjectInfo(object_type=DIRECTORY, object_id=sha1_git)) + swh_objects.append( + SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=sha1_git) + ) vault_cooking.update( {"directory_context": True, "directory_swhid": f"swh:1:dir:{sha1_git}",} ) if revision_found: - swh_objects.append(SWHObjectInfo(object_type=REVISION, object_id=revision_id)) + swh_objects.append( + SWHObjectInfo(object_type=ObjectType.REVISION, object_id=revision_id) + ) vault_cooking.update( {"revision_context": True, "revision_swhid": f"swh:1:rev:{revision_id}",} ) - swh_objects.append(SWHObjectInfo(object_type=SNAPSHOT, object_id=snapshot_id)) + swh_objects.append( + SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id) + ) visit_date = None visit_type = None if visit_info: visit_date = format_utc_iso_date(visit_info["date"]) visit_type = visit_info["type"] release_id = snapshot_context["release_id"] if release_id: - swh_objects.append(SWHObjectInfo(object_type=RELEASE, object_id=release_id)) + swh_objects.append( + SWHObjectInfo(object_type=ObjectType.RELEASE, object_id=release_id) + ) dir_metadata = DirectoryMetadata( - object_type=DIRECTORY, + object_type=ObjectType.DIRECTORY, object_id=sha1_git, directory=sha1_git, nb_files=nb_files, nb_dirs=nb_dirs, sum_file_sizes=sum_file_sizes, root_directory=root_directory, path=dir_path, revision=revision_id, revision_found=revision_found, release=release_id, snapshot=snapshot_id, origin_url=origin_url, visit_date=visit_date, visit_type=visit_type, ) swhids_info = get_swhids_info(swh_objects, snapshot_context, dir_metadata) dir_path = "/".join([bc["name"] for bc in breadcrumbs]) + "/" context_found = "snapshot: %s" % snapshot_context["snapshot_id"] if origin_info: context_found = "origin: %s" % origin_info["url"] heading = "Directory - %s - %s - %s" % ( dir_path, snapshot_context["branch"], context_found, ) top_right_link = None if not snapshot_context["is_empty"] and revision_found: top_right_link = { "url": history_url, "icon": swh_object_icons["revisions history"], "text": "History", } return render( request, "browse/directory.html", { "heading": heading, "swh_object_name": "Directory", "swh_object_metadata": dir_metadata, "dirs": dirs, "files": files, "breadcrumbs": breadcrumbs if root_directory else [], "top_right_link": top_right_link, "readme_name": readme_name, "readme_url": readme_url, "readme_html": readme_html, "snapshot_context": snapshot_context, "vault_cooking": vault_cooking, "show_actions": True, "swhids_info": swhids_info, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], }, status=error_info["status_code"], ) def browse_snapshot_content( request, snapshot_id=None, origin_url=None, timestamp=None, path=None, selected_language=None, ): """ Django view implementation for browsing a content in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) if path is None: raise BadInputExc("The path of a content must be given as query parameter.") snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), path=path, browse_context="content", branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), ) root_directory = snapshot_context["root_directory"] sha1_git = None query_string = None content_data = {} directory_id = None split_path = path.split("/") filename = split_path[-1] filepath = path[: -len(filename)] error_info = { "status_code": 200, "description": None, } if root_directory: try: content_info = archive.lookup_directory_with_path(root_directory, path) sha1_git = content_info["target"] query_string = "sha1_git:" + sha1_git content_data = request_content(query_string) if filepath: dir_info = archive.lookup_directory_with_path(root_directory, filepath) directory_id = dir_info["target"] else: directory_id = root_directory except NotFoundExc as e: error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" revision_id = snapshot_context["revision_id"] origin_info = snapshot_context["origin_info"] visit_info = snapshot_context["visit_info"] snapshot_id = snapshot_context["snapshot_id"] if content_data.get("raw_data") is not None: content_display_data = prepare_content_for_display( content_data["raw_data"], content_data["mimetype"], path ) content_data.update(content_display_data) # Override language with user-selected language if selected_language is not None: content_data["language"] = selected_language available_languages = None if content_data.get("mimetype") is not None and "text/" in content_data["mimetype"]: available_languages = highlightjs.get_supported_languages() breadcrumbs = _build_breadcrumbs(snapshot_context, filepath) breadcrumbs.append({"name": filename, "url": None}) browse_content_link = gen_content_link(sha1_git) content_raw_url = None if query_string: content_raw_url = reverse( "browse-content-raw", url_args={"query_string": query_string}, query_params={"filename": filename}, ) content_checksums = content_data.get("checksums", {}) swh_objects = [ - SWHObjectInfo(object_type=CONTENT, object_id=content_checksums.get("sha1_git")), - SWHObjectInfo(object_type=DIRECTORY, object_id=directory_id), - SWHObjectInfo(object_type=REVISION, object_id=revision_id), - SWHObjectInfo(object_type=SNAPSHOT, object_id=snapshot_id), + SWHObjectInfo( + object_type=ObjectType.CONTENT, object_id=content_checksums.get("sha1_git") + ), + SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=directory_id), + SWHObjectInfo(object_type=ObjectType.REVISION, object_id=revision_id), + SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id), ] visit_date = None visit_type = None if visit_info: visit_date = format_utc_iso_date(visit_info["date"]) visit_type = visit_info["type"] release_id = snapshot_context["release_id"] if release_id: - swh_objects.append(SWHObjectInfo(object_type=RELEASE, object_id=release_id)) + swh_objects.append( + SWHObjectInfo(object_type=ObjectType.RELEASE, object_id=release_id) + ) content_metadata = ContentMetadata( - object_type=CONTENT, + object_type=ObjectType.CONTENT, object_id=content_checksums.get("sha1_git"), sha1=content_checksums.get("sha1"), sha1_git=content_checksums.get("sha1_git"), sha256=content_checksums.get("sha256"), blake2s256=content_checksums.get("blake2s256"), content_url=browse_content_link, mimetype=content_data.get("mimetype"), encoding=content_data.get("encoding"), size=content_data.get("length", 0), language=content_data.get("language"), root_directory=root_directory, path=f"/{filepath}", filename=filename, directory=directory_id, revision=revision_id, release=release_id, snapshot=snapshot_id, origin_url=origin_url, visit_date=visit_date, visit_type=visit_type, ) swhids_info = get_swhids_info(swh_objects, snapshot_context, content_metadata) content_path = "/".join([bc["name"] for bc in breadcrumbs]) context_found = "snapshot: %s" % snapshot_context["snapshot_id"] if origin_info: context_found = "origin: %s" % origin_info["url"] heading = "Content - %s - %s - %s" % ( content_path, snapshot_context["branch"], context_found, ) top_right_link = None if not snapshot_context["is_empty"]: top_right_link = { "url": content_raw_url, "icon": swh_object_icons["content"], "text": "Raw File", } return render( request, "browse/content.html", { "heading": heading, "swh_object_name": "Content", "swh_object_metadata": content_metadata, "content": content_data.get("content_data"), "content_size": content_data.get("length"), "max_content_size": content_display_max_size, "filename": filename, "encoding": content_data.get("encoding"), "mimetype": content_data.get("mimetype"), "language": content_data.get("language"), "available_languages": available_languages, "breadcrumbs": breadcrumbs if root_directory else [], "top_right_link": top_right_link, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": True, "swhids_info": swhids_info, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], }, status=error_info["status_code"], ) PER_PAGE = 100 def browse_snapshot_log(request, snapshot_id=None, origin_url=None, timestamp=None): """ Django view implementation for browsing a revision history in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), browse_context="log", branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), ) revision_id = snapshot_context["revision_id"] per_page = int(request.GET.get("per_page", PER_PAGE)) offset = int(request.GET.get("offset", 0)) revs_ordering = request.GET.get("revs_ordering", "committer_date") session_key = "rev_%s_log_ordering_%s" % (revision_id, revs_ordering) rev_log_session = request.session.get(session_key, None) rev_log = [] revs_walker_state = None if rev_log_session: rev_log = rev_log_session["rev_log"] revs_walker_state = rev_log_session["revs_walker_state"] if len(rev_log) < offset + per_page: revs_walker = archive.get_revisions_walker( revs_ordering, revision_id, max_revs=offset + per_page + 1, state=revs_walker_state, ) rev_log += [rev["id"] for rev in revs_walker] revs_walker_state = revs_walker.export_state() revs = rev_log[offset : offset + per_page] revision_log = archive.lookup_revision_multiple(revs) request.session[session_key] = { "rev_log": rev_log, "revs_walker_state": revs_walker_state, } origin_info = snapshot_context["origin_info"] visit_info = snapshot_context["visit_info"] url_args = snapshot_context["url_args"] query_params = snapshot_context["query_params"] snapshot_id = snapshot_context["snapshot_id"] query_params["per_page"] = per_page revs_ordering = request.GET.get("revs_ordering", "") query_params["revs_ordering"] = revs_ordering or None if origin_info: browse_view_name = "browse-origin-log" else: browse_view_name = "browse-snapshot-log" prev_log_url = None if len(rev_log) > offset + per_page: query_params["offset"] = offset + per_page prev_log_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) next_log_url = None if offset != 0: query_params["offset"] = offset - per_page next_log_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) revision_log_data = format_log_entries(revision_log, per_page, snapshot_context) browse_rev_link = gen_revision_link(revision_id) browse_log_link = gen_revision_log_link(revision_id) browse_snp_link = gen_snapshot_link(snapshot_id) revision_metadata = { "context-independent revision": browse_rev_link, "context-independent revision history": browse_log_link, "context-independent snapshot": browse_snp_link, "snapshot": snapshot_id, } if origin_info: revision_metadata["origin url"] = origin_info["url"] revision_metadata["origin visit date"] = format_utc_iso_date(visit_info["date"]) revision_metadata["origin visit type"] = visit_info["type"] swh_objects = [ - SWHObjectInfo(object_type=REVISION, object_id=revision_id), - SWHObjectInfo(object_type=SNAPSHOT, object_id=snapshot_id), + SWHObjectInfo(object_type=ObjectType.REVISION, object_id=revision_id), + SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id), ] release_id = snapshot_context["release_id"] if release_id: - swh_objects.append(SWHObjectInfo(object_type=RELEASE, object_id=release_id)) + swh_objects.append( + SWHObjectInfo(object_type=ObjectType.RELEASE, object_id=release_id) + ) browse_rel_link = gen_release_link(release_id) revision_metadata["release"] = release_id revision_metadata["context-independent release"] = browse_rel_link swhids_info = get_swhids_info(swh_objects, snapshot_context) context_found = "snapshot: %s" % snapshot_context["snapshot_id"] if origin_info: context_found = "origin: %s" % origin_info["url"] heading = "Revision history - %s - %s" % (snapshot_context["branch"], context_found) return render( request, "browse/revision-log.html", { "heading": heading, "swh_object_name": "Revisions history", "swh_object_metadata": revision_metadata, "revision_log": revision_log_data, "revs_ordering": revs_ordering, "next_log_url": next_log_url, "prev_log_url": prev_log_url, "breadcrumbs": None, "top_right_link": None, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": True, "swhids_info": swhids_info, }, ) def browse_snapshot_branches( request, snapshot_id=None, origin_url=None, timestamp=None, branch_name_include=None ): """ Django view implementation for browsing a list of branches in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), ) branches_bc = request.GET.get("branches_breadcrumbs", "") branches_bc = branches_bc.split(",") if branches_bc else [] branches_from = branches_bc[-1] if branches_bc else "" origin_info = snapshot_context["origin_info"] url_args = snapshot_context["url_args"] query_params = snapshot_context["query_params"] if origin_info: browse_view_name = "browse-origin-directory" else: browse_view_name = "browse-snapshot-directory" snapshot = archive.lookup_snapshot( snapshot_context["snapshot_id"], branches_from, PER_PAGE + 1, target_types=["revision", "alias"], branch_name_include_substring=branch_name_include, ) displayed_branches = [] if snapshot: displayed_branches, _, _ = process_snapshot_branches(snapshot) for branch in displayed_branches: rev_query_params = {} if origin_info: rev_query_params["origin_url"] = origin_info["url"] revision_url = reverse( "browse-revision", url_args={"sha1_git": branch["revision"]}, query_params=query_params, ) query_params["branch"] = branch["name"] directory_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) del query_params["branch"] branch["revision_url"] = revision_url branch["directory_url"] = directory_url if origin_info: browse_view_name = "browse-origin-branches" else: browse_view_name = "browse-snapshot-branches" prev_branches_url = None next_branches_url = None if branches_bc: query_params_prev = dict(query_params) query_params_prev["branches_breadcrumbs"] = ",".join(branches_bc[:-1]) prev_branches_url = reverse( browse_view_name, url_args=url_args, query_params=query_params_prev ) elif branches_from: prev_branches_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) if snapshot and snapshot["next_branch"] is not None: query_params_next = dict(query_params) next_branch = displayed_branches[-1]["name"] del displayed_branches[-1] branches_bc.append(next_branch) query_params_next["branches_breadcrumbs"] = ",".join(branches_bc) next_branches_url = reverse( browse_view_name, url_args=url_args, query_params=query_params_next ) heading = "Branches - " if origin_info: heading += "origin: %s" % origin_info["url"] else: heading += "snapshot: %s" % snapshot_id return render( request, "browse/branches.html", { "heading": heading, "swh_object_name": "Branches", "swh_object_metadata": {}, "top_right_link": None, "displayed_branches": displayed_branches, "prev_branches_url": prev_branches_url, "next_branches_url": next_branches_url, "snapshot_context": snapshot_context, "search_string": branch_name_include or "", }, ) def browse_snapshot_releases( request, snapshot_id=None, origin_url=None, timestamp=None, release_name_include=None, ): """ Django view implementation for browsing a list of releases in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), ) rel_bc = request.GET.get("releases_breadcrumbs", "") rel_bc = rel_bc.split(",") if rel_bc else [] rel_from = rel_bc[-1] if rel_bc else "" origin_info = snapshot_context["origin_info"] url_args = snapshot_context["url_args"] query_params = snapshot_context["query_params"] snapshot = archive.lookup_snapshot( snapshot_context["snapshot_id"], rel_from, PER_PAGE + 1, target_types=["release", "alias"], branch_name_include_substring=release_name_include, ) displayed_releases = [] if snapshot: _, displayed_releases, _ = process_snapshot_branches(snapshot) for release in displayed_releases: query_params_tgt = {"snapshot": snapshot_id} if origin_info: query_params_tgt["origin_url"] = origin_info["url"] release_url = reverse( "browse-release", url_args={"sha1_git": release["id"]}, query_params=query_params_tgt, ) target_url = "" tooltip = ( f"The release {release['name']} targets " f"{release['target_type']} {release['target']}" ) if release["target_type"] == "revision": target_url = reverse( "browse-revision", url_args={"sha1_git": release["target"]}, query_params=query_params_tgt, ) elif release["target_type"] == "directory": target_url = reverse( "browse-directory", url_args={"sha1_git": release["target"]}, query_params=query_params_tgt, ) elif release["target_type"] == "content": target_url = reverse( "browse-content", url_args={"query_string": release["target"]}, query_params=query_params_tgt, ) elif release["target_type"] == "release": target_url = reverse( "browse-release", url_args={"sha1_git": release["target"]}, query_params=query_params_tgt, ) tooltip = ( f"The release {release['name']} " f"is an alias for release {release['target']}" ) release["release_url"] = release_url release["target_url"] = target_url release["tooltip"] = tooltip if origin_info: browse_view_name = "browse-origin-releases" else: browse_view_name = "browse-snapshot-releases" prev_releases_url = None next_releases_url = None if rel_bc: query_params_prev = dict(query_params) query_params_prev["releases_breadcrumbs"] = ",".join(rel_bc[:-1]) prev_releases_url = reverse( browse_view_name, url_args=url_args, query_params=query_params_prev ) elif rel_from: prev_releases_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) if snapshot and snapshot["next_branch"] is not None: query_params_next = dict(query_params) next_rel = displayed_releases[-1]["branch_name"] del displayed_releases[-1] rel_bc.append(next_rel) query_params_next["releases_breadcrumbs"] = ",".join(rel_bc) next_releases_url = reverse( browse_view_name, url_args=url_args, query_params=query_params_next ) heading = "Releases - " if origin_info: heading += "origin: %s" % origin_info["url"] else: heading += "snapshot: %s" % snapshot_id return render( request, "browse/releases.html", { "heading": heading, "top_panel_visible": False, "top_panel_collapsible": False, "swh_object_name": "Releases", "swh_object_metadata": {}, "top_right_link": None, "displayed_releases": displayed_releases, "prev_releases_url": prev_releases_url, "next_releases_url": next_releases_url, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": False, "search_string": release_name_include or "", }, ) diff --git a/swh/web/browse/views/content.py b/swh/web/browse/views/content.py index a8dfbcd8..497eba0f 100644 --- a/swh/web/browse/views/content.py +++ b/swh/web/browse/views/content.py @@ -1,394 +1,401 @@ -# Copyright (C) 2017-2020 The Software Heritage developers +# Copyright (C) 2017-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import difflib from distutils.util import strtobool import sentry_sdk from django.http import HttpResponse, JsonResponse from django.shortcuts import render from swh.model.hashutil import hash_to_hex -from swh.model.identifiers import CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT +from swh.model.identifiers import ObjectType from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import ( content_display_max_size, gen_link, prepare_content_for_display, request_content, ) from swh.web.common import archive, highlightjs, query from swh.web.common.exc import NotFoundExc, http_status_code_message from swh.web.common.identifiers import get_swhids_info from swh.web.common.typing import ContentMetadata, SWHObjectInfo from swh.web.common.utils import gen_path_info, reverse, swh_object_icons @browse_route( r"content/(?P[0-9a-z_:]*[0-9a-f]+.)/raw/", view_name="browse-content-raw", checksum_args=["query_string"], ) def content_raw(request, query_string): """Django view that produces a raw display of a content identified by its hash value. The url that points to it is :http:get:`/browse/content/[(algo_hash):](hash)/raw/` """ re_encode = bool(strtobool(request.GET.get("re_encode", "false"))) algo, checksum = query.parse_hash(query_string) checksum = hash_to_hex(checksum) content_data = request_content(query_string, max_size=None, re_encode=re_encode) filename = request.GET.get("filename", None) if not filename: filename = "%s_%s" % (algo, checksum) if ( content_data["mimetype"].startswith("text/") or content_data["mimetype"] == "inode/x-empty" ): response = HttpResponse(content_data["raw_data"], content_type="text/plain") response["Content-disposition"] = "filename=%s" % filename else: response = HttpResponse( content_data["raw_data"], content_type="application/octet-stream" ) response["Content-disposition"] = "attachment; filename=%s" % filename return response _auto_diff_size_limit = 20000 @browse_route( r"content/(?P.*)/diff/(?P.*)", view_name="diff-contents", ) def _contents_diff(request, from_query_string, to_query_string): """ Browse endpoint used to compute unified diffs between two contents. Diffs are generated only if the two contents are textual. By default, diffs whose size are greater than 20 kB will not be generated. To force the generation of large diffs, the 'force' boolean query parameter must be used. Args: request: input django http request from_query_string: a string of the form "[ALGO_HASH:]HASH" where optional ALGO_HASH can be either ``sha1``, ``sha1_git``, ``sha256``, or ``blake2s256`` (default to ``sha1``) and HASH the hexadecimal representation of the hash value identifying the first content to_query_string: same as above for identifying the second content Returns: A JSON object containing the unified diff. """ diff_data = {} content_from = None content_to = None content_from_size = 0 content_to_size = 0 content_from_lines = [] content_to_lines = [] force = request.GET.get("force", "false") path = request.GET.get("path", None) language = "nohighlight" force = bool(strtobool(force)) if from_query_string == to_query_string: diff_str = "File renamed without changes" else: try: text_diff = True if from_query_string: content_from = request_content(from_query_string, max_size=None) content_from_display_data = prepare_content_for_display( content_from["raw_data"], content_from["mimetype"], path ) language = content_from_display_data["language"] content_from_size = content_from["length"] if not ( content_from["mimetype"].startswith("text/") or content_from["mimetype"] == "inode/x-empty" ): text_diff = False if text_diff and to_query_string: content_to = request_content(to_query_string, max_size=None) content_to_display_data = prepare_content_for_display( content_to["raw_data"], content_to["mimetype"], path ) language = content_to_display_data["language"] content_to_size = content_to["length"] if not ( content_to["mimetype"].startswith("text/") or content_to["mimetype"] == "inode/x-empty" ): text_diff = False diff_size = abs(content_to_size - content_from_size) if not text_diff: diff_str = "Diffs are not generated for non textual content" language = "nohighlight" elif not force and diff_size > _auto_diff_size_limit: diff_str = "Large diffs are not automatically computed" language = "nohighlight" else: if content_from: content_from_lines = ( content_from["raw_data"].decode("utf-8").splitlines(True) ) if content_from_lines and content_from_lines[-1][-1] != "\n": content_from_lines[-1] += "[swh-no-nl-marker]\n" if content_to: content_to_lines = ( content_to["raw_data"].decode("utf-8").splitlines(True) ) if content_to_lines and content_to_lines[-1][-1] != "\n": content_to_lines[-1] += "[swh-no-nl-marker]\n" diff_lines = difflib.unified_diff(content_from_lines, content_to_lines) diff_str = "".join(list(diff_lines)[2:]) except Exception as exc: sentry_sdk.capture_exception(exc) diff_str = str(exc) diff_data["diff_str"] = diff_str diff_data["language"] = language return JsonResponse(diff_data) @browse_route( r"content/(?P[0-9a-z_:]*[0-9a-f]+.)/", view_name="browse-content", checksum_args=["query_string"], ) def content_display(request, query_string): """Django view that produces an HTML display of a content identified by its hash value. The url that points to it is :http:get:`/browse/content/[(algo_hash):](hash)/` """ algo, checksum = query.parse_hash(query_string) checksum = hash_to_hex(checksum) origin_url = request.GET.get("origin_url") selected_language = request.GET.get("language") if not origin_url: origin_url = request.GET.get("origin") snapshot_id = request.GET.get("snapshot") path = request.GET.get("path") content_data = {} error_info = {"status_code": 200, "description": None} try: content_data = request_content(query_string) except NotFoundExc as e: error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" snapshot_context = None if origin_url is not None or snapshot_id is not None: try: snapshot_context = get_snapshot_context( origin_url=origin_url, snapshot_id=snapshot_id, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), path=path, - browse_context=CONTENT, + browse_context="content", ) except NotFoundExc as e: if str(e).startswith("Origin"): raw_cnt_url = reverse( "browse-content", url_args={"query_string": query_string} ) error_message = ( "The Software Heritage archive has a content " "with the hash you provided but the origin " "mentioned in your request appears broken: %s. " "Please check the URL and try again.\n\n" "Nevertheless, you can still browse the content " "without origin information: %s" % (gen_link(origin_url), gen_link(raw_cnt_url)) ) raise NotFoundExc(error_message) else: raise e content = None language = None mimetype = None if content_data.get("raw_data") is not None: content_display_data = prepare_content_for_display( content_data["raw_data"], content_data["mimetype"], path ) content = content_display_data["content_data"] language = content_display_data["language"] mimetype = content_display_data["mimetype"] # Override language with user-selected language if selected_language is not None: language = selected_language available_languages = None if mimetype and "text/" in mimetype: available_languages = highlightjs.get_supported_languages() filename = None path_info = None directory_id = None root_dir = None if snapshot_context: root_dir = snapshot_context.get("root_directory") query_params = snapshot_context["query_params"] if snapshot_context else {} breadcrumbs = [] if path: split_path = path.split("/") root_dir = root_dir or split_path[0] filename = split_path[-1] if root_dir != path: path = path.replace(root_dir + "/", "") path = path[: -len(filename)] path_info = gen_path_info(path) query_params.pop("path", None) dir_url = reverse( "browse-directory", url_args={"sha1_git": root_dir}, query_params=query_params, ) breadcrumbs.append({"name": root_dir[:7], "url": dir_url}) for pi in path_info: query_params["path"] = pi["path"] dir_url = reverse( "browse-directory", url_args={"sha1_git": root_dir}, query_params=query_params, ) breadcrumbs.append({"name": pi["name"], "url": dir_url}) breadcrumbs.append({"name": filename, "url": None}) if path and root_dir != path: dir_info = archive.lookup_directory_with_path(root_dir, path) directory_id = dir_info["target"] elif root_dir != path: directory_id = root_dir else: root_dir = None query_params = {"filename": filename} content_checksums = content_data.get("checksums", {}) content_url = reverse("browse-content", url_args={"query_string": query_string},) content_raw_url = reverse( "browse-content-raw", url_args={"query_string": query_string}, query_params=query_params, ) content_metadata = ContentMetadata( - object_type=CONTENT, + object_type=ObjectType.CONTENT, object_id=content_checksums.get("sha1_git"), sha1=content_checksums.get("sha1"), sha1_git=content_checksums.get("sha1_git"), sha256=content_checksums.get("sha256"), blake2s256=content_checksums.get("blake2s256"), content_url=content_url, mimetype=content_data.get("mimetype"), encoding=content_data.get("encoding"), size=content_data.get("length", 0), language=content_data.get("language"), root_directory=root_dir, path=f"/{path}" if path else None, filename=filename or "", directory=directory_id, revision=None, release=None, snapshot=None, origin_url=origin_url, ) swh_objects = [ - SWHObjectInfo(object_type=CONTENT, object_id=content_checksums.get("sha1_git")) + SWHObjectInfo( + object_type=ObjectType.CONTENT, object_id=content_checksums.get("sha1_git") + ) ] if directory_id: - swh_objects.append(SWHObjectInfo(object_type=DIRECTORY, object_id=directory_id)) + swh_objects.append( + SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=directory_id) + ) if snapshot_context: swh_objects.append( SWHObjectInfo( - object_type=REVISION, object_id=snapshot_context["revision_id"] + object_type=ObjectType.REVISION, + object_id=snapshot_context["revision_id"], ) ) swh_objects.append( SWHObjectInfo( - object_type=SNAPSHOT, object_id=snapshot_context["snapshot_id"] + object_type=ObjectType.SNAPSHOT, + object_id=snapshot_context["snapshot_id"], ) ) if snapshot_context["release_id"]: swh_objects.append( SWHObjectInfo( - object_type=RELEASE, object_id=snapshot_context["release_id"] + object_type=ObjectType.RELEASE, + object_id=snapshot_context["release_id"], ) ) swhids_info = get_swhids_info( swh_objects, snapshot_context, extra_context=content_metadata, ) heading = "Content - %s" % content_checksums.get("sha1_git") if breadcrumbs: content_path = "/".join([bc["name"] for bc in breadcrumbs]) heading += " - %s" % content_path return render( request, "browse/content.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"], "swh_object_name": "Content", "swh_object_metadata": content_metadata, "content": content, "content_size": content_data.get("length"), "max_content_size": content_display_max_size, "filename": filename, "encoding": content_data.get("encoding"), "mimetype": mimetype, "language": language, "available_languages": available_languages, "breadcrumbs": breadcrumbs, "top_right_link": { "url": content_raw_url, "icon": swh_object_icons["content"], "text": "Raw File", }, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": True, "swhids_info": swhids_info, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], }, status=error_info["status_code"], ) diff --git a/swh/web/browse/views/directory.py b/swh/web/browse/views/directory.py index ba5fc960..56b7654e 100644 --- a/swh/web/browse/views/directory.py +++ b/swh/web/browse/views/directory.py @@ -1,278 +1,281 @@ -# Copyright (C) 2017-2020 The Software Heritage developers +# Copyright (C) 2017-2021 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 os import sentry_sdk from django.http import HttpResponse from django.shortcuts import redirect, render -from swh.model.identifiers import DIRECTORY, RELEASE, REVISION, SNAPSHOT +from swh.model.identifiers import ObjectType from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import gen_link, get_directory_entries, get_readme_to_display from swh.web.common import archive from swh.web.common.exc import NotFoundExc, http_status_code_message from swh.web.common.identifiers import get_swhids_info from swh.web.common.typing import DirectoryMetadata, SWHObjectInfo from swh.web.common.utils import gen_path_info, reverse, swh_object_icons def _directory_browse(request, sha1_git, path=None): root_sha1_git = sha1_git error_info = {"status_code": 200, "description": None} if path: try: dir_info = archive.lookup_directory_with_path(sha1_git, path) sha1_git = dir_info["target"] except NotFoundExc as e: error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" sha1_git = None dirs, files = [], [] if sha1_git is not None: dirs, files = get_directory_entries(sha1_git) origin_url = request.GET.get("origin_url") if not origin_url: origin_url = request.GET.get("origin") snapshot_id = request.GET.get("snapshot") snapshot_context = None if origin_url is not None or snapshot_id is not None: try: snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), path=path, ) except NotFoundExc as e: if str(e).startswith("Origin"): raw_dir_url = reverse( "browse-directory", url_args={"sha1_git": sha1_git} ) error_message = ( "The Software Heritage archive has a directory " "with the hash you provided but the origin " "mentioned in your request appears broken: %s. " "Please check the URL and try again.\n\n" "Nevertheless, you can still browse the directory " "without origin information: %s" % (gen_link(origin_url), gen_link(raw_dir_url)) ) raise NotFoundExc(error_message) else: raise e path_info = gen_path_info(path) query_params = snapshot_context["query_params"] if snapshot_context else {} breadcrumbs = [] breadcrumbs.append( { "name": root_sha1_git[:7], "url": reverse( "browse-directory", url_args={"sha1_git": root_sha1_git}, query_params={**query_params, "path": None}, ), } ) for pi in path_info: breadcrumbs.append( { "name": pi["name"], "url": reverse( "browse-directory", url_args={"sha1_git": root_sha1_git}, query_params={**query_params, "path": pi["path"],}, ), } ) path = "" if path is None else (path + "/") for d in dirs: if d["type"] == "rev": d["url"] = reverse( "browse-revision", url_args={"sha1_git": d["target"]}, query_params=query_params, ) else: d["url"] = reverse( "browse-directory", url_args={"sha1_git": root_sha1_git}, query_params={**query_params, "path": path + d["name"],}, ) sum_file_sizes = 0 readmes = {} for f in files: query_string = "sha1_git:" + f["target"] f["url"] = reverse( "browse-content", url_args={"query_string": query_string}, query_params={ **query_params, "path": root_sha1_git + "/" + path + f["name"], }, ) if f["length"] is not None: sum_file_sizes += f["length"] if f["name"].lower().startswith("readme"): readmes[f["name"]] = f["checksums"]["sha1"] readme_name, readme_url, readme_html = get_readme_to_display(readmes) dir_metadata = DirectoryMetadata( - object_type=DIRECTORY, + object_type=ObjectType.DIRECTORY, object_id=sha1_git, directory=root_sha1_git, nb_files=len(files), nb_dirs=len(dirs), sum_file_sizes=sum_file_sizes, root_directory=root_sha1_git, path=f"/{path}" if path else None, revision=None, revision_found=None, release=None, snapshot=None, ) vault_cooking = { "directory_context": True, "directory_swhid": f"swh:1:dir:{sha1_git}", "revision_context": False, "revision_swhid": None, } - swh_objects = [SWHObjectInfo(object_type=DIRECTORY, object_id=sha1_git)] + swh_objects = [SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=sha1_git)] if snapshot_context: swh_objects.append( SWHObjectInfo( - object_type=REVISION, object_id=snapshot_context["revision_id"] + object_type=ObjectType.REVISION, + object_id=snapshot_context["revision_id"], ) ) swh_objects.append( SWHObjectInfo( - object_type=SNAPSHOT, object_id=snapshot_context["snapshot_id"] + object_type=ObjectType.SNAPSHOT, + object_id=snapshot_context["snapshot_id"], ) ) if snapshot_context["release_id"]: swh_objects.append( SWHObjectInfo( - object_type=RELEASE, object_id=snapshot_context["release_id"] + object_type=ObjectType.RELEASE, + object_id=snapshot_context["release_id"], ) ) swhids_info = get_swhids_info(swh_objects, snapshot_context, dir_metadata) heading = "Directory - %s" % sha1_git if breadcrumbs: dir_path = "/".join([bc["name"] for bc in breadcrumbs]) + "/" heading += " - %s" % dir_path top_right_link = None if snapshot_context is not None and not snapshot_context["is_empty"]: history_url = reverse( "browse-revision-log", url_args={"sha1_git": snapshot_context["revision_id"]}, query_params=query_params, ) top_right_link = { "url": history_url, "icon": swh_object_icons["revisions history"], "text": "History", } return render( request, "browse/directory.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"], "swh_object_name": "Directory", "swh_object_metadata": dir_metadata, "dirs": dirs, "files": files, "breadcrumbs": breadcrumbs, "top_right_link": top_right_link, "readme_name": readme_name, "readme_url": readme_url, "readme_html": readme_html, "snapshot_context": snapshot_context, "vault_cooking": vault_cooking, "show_actions": True, "swhids_info": swhids_info, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], }, status=error_info["status_code"], ) @browse_route( r"directory/(?P[0-9a-f]+)/", view_name="browse-directory", checksum_args=["sha1_git"], ) def directory_browse(request, sha1_git): """Django view for browsing the content of a directory identified by its sha1_git value. The url that points to it is :http:get:`/browse/directory/(sha1_git)/` """ return _directory_browse(request, sha1_git, request.GET.get("path")) @browse_route( r"directory/(?P[0-9a-f]+)/(?P.+)/", view_name="browse-directory-legacy", checksum_args=["sha1_git"], ) def directory_browse_legacy(request, sha1_git, path): """Django view for browsing the content of a directory identified by its sha1_git value. The url that points to it is :http:get:`/browse/directory/(sha1_git)/(path)/` """ return _directory_browse(request, sha1_git, path) @browse_route( r"directory/resolve/content-path/(?P[0-9a-f]+)/", view_name="browse-directory-resolve-content-path", checksum_args=["sha1_git"], ) def _directory_resolve_content_path(request, sha1_git): """ Internal endpoint redirecting to data url for a specific file path relative to a root directory. """ try: path = os.path.normpath(request.GET.get("path")) if not path.startswith("../"): dir_info = archive.lookup_directory_with_path(sha1_git, path) if dir_info["type"] == "file": sha1 = dir_info["checksums"]["sha1"] data_url = reverse( "browse-content-raw", url_args={"query_string": sha1} ) return redirect(data_url) except Exception as exc: sentry_sdk.capture_exception(exc) return HttpResponse(status=404) diff --git a/swh/web/browse/views/release.py b/swh/web/browse/views/release.py index 8edc000c..9ff72c83 100644 --- a/swh/web/browse/views/release.py +++ b/swh/web/browse/views/release.py @@ -1,239 +1,245 @@ -# Copyright (C) 2017-2020 The Software Heritage developers +# Copyright (C) 2017-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import sentry_sdk from django.shortcuts import render -from swh.model.identifiers import CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT +from swh.model.identifiers import ObjectType from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import ( gen_content_link, gen_directory_link, gen_link, gen_person_mail_link, gen_release_link, gen_revision_link, ) from swh.web.common import archive from swh.web.common.exc import NotFoundExc from swh.web.common.identifiers import get_swhids_info from swh.web.common.typing import ReleaseMetadata, SWHObjectInfo from swh.web.common.utils import format_utc_iso_date, reverse @browse_route( r"release/(?P[0-9a-f]+)/", view_name="browse-release", checksum_args=["sha1_git"], ) def release_browse(request, sha1_git): """ Django view that produces an HTML display of a release identified by its id. The url that points to it is :http:get:`/browse/release/(sha1_git)/`. """ release = archive.lookup_release(sha1_git) snapshot_context = {} origin_info = None snapshot_id = request.GET.get("snapshot_id") if not snapshot_id: snapshot_id = request.GET.get("snapshot") origin_url = request.GET.get("origin_url") if not origin_url: origin_url = request.GET.get("origin") timestamp = request.GET.get("timestamp") visit_id = request.GET.get("visit_id") if origin_url: try: snapshot_context = get_snapshot_context( snapshot_id, origin_url, timestamp, visit_id, release_name=release["name"], ) except NotFoundExc as e: raw_rel_url = reverse("browse-release", url_args={"sha1_git": sha1_git}) error_message = ( "The Software Heritage archive has a release " "with the hash you provided but the origin " "mentioned in your request appears broken: %s. " "Please check the URL and try again.\n\n" "Nevertheless, you can still browse the release " "without origin information: %s" % (gen_link(origin_url), gen_link(raw_rel_url)) ) if str(e).startswith("Origin"): raise NotFoundExc(error_message) else: raise e origin_info = snapshot_context["origin_info"] elif snapshot_id: snapshot_context = get_snapshot_context( snapshot_id, release_name=release["name"] ) snapshot_id = snapshot_context.get("snapshot_id", None) release_metadata = ReleaseMetadata( - object_type=RELEASE, + object_type=ObjectType.RELEASE, object_id=sha1_git, release=sha1_git, author=release["author"]["fullname"] if release["author"] else "None", author_url=gen_person_mail_link(release["author"]) if release["author"] else "None", date=format_utc_iso_date(release["date"]), name=release["name"], synthetic=release["synthetic"], target=release["target"], target_type=release["target_type"], snapshot=snapshot_id, origin_url=origin_url, ) release_note_lines = [] if release["message"]: release_note_lines = release["message"].split("\n") - swh_objects = [SWHObjectInfo(object_type=RELEASE, object_id=sha1_git)] + swh_objects = [SWHObjectInfo(object_type=ObjectType.RELEASE, object_id=sha1_git)] vault_cooking = None rev_directory = None target_link = None - if release["target_type"] == REVISION: + if release["target_type"] == ObjectType.REVISION.name.lower(): target_link = gen_revision_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) try: revision = archive.lookup_revision(release["target"]) rev_directory = revision["directory"] vault_cooking = { "directory_context": True, "directory_swhid": f"swh:1:dir:{rev_directory}", "revision_context": True, "revision_swhid": f"swh:1:rev:{release['target']}", } swh_objects.append( - SWHObjectInfo(object_type=REVISION, object_id=release["target"]) + SWHObjectInfo( + object_type=ObjectType.REVISION, object_id=release["target"] + ) ) swh_objects.append( - SWHObjectInfo(object_type=DIRECTORY, object_id=rev_directory) + SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=rev_directory) ) except Exception as exc: sentry_sdk.capture_exception(exc) - elif release["target_type"] == DIRECTORY: + elif release["target_type"] == ObjectType.DIRECTORY.name.lower(): target_link = gen_directory_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) try: # check directory exists archive.lookup_directory(release["target"]) vault_cooking = { "directory_context": True, "directory_swhid": f"swh:1:dir:{release['target']}", "revision_context": False, "revision_swhid": None, } swh_objects.append( - SWHObjectInfo(object_type=DIRECTORY, object_id=release["target"]) + SWHObjectInfo( + object_type=ObjectType.DIRECTORY, object_id=release["target"] + ) ) except Exception as exc: sentry_sdk.capture_exception(exc) - elif release["target_type"] == CONTENT: + elif release["target_type"] == ObjectType.CONTENT.name.lower(): target_link = gen_content_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) swh_objects.append( - SWHObjectInfo(object_type=CONTENT, object_id=release["target"]) + SWHObjectInfo(object_type=ObjectType.CONTENT, object_id=release["target"]) ) - elif release["target_type"] == RELEASE: + elif release["target_type"] == ObjectType.RELEASE.name.lower(): target_link = gen_release_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) rev_directory_url = None if rev_directory is not None: if origin_info: rev_directory_url = reverse( "browse-origin-directory", query_params={ "origin_url": origin_info["url"], "release": release["name"], "snapshot": snapshot_id, }, ) elif snapshot_id: rev_directory_url = reverse( "browse-snapshot-directory", url_args={"snapshot_id": snapshot_id}, query_params={"release": release["name"]}, ) else: rev_directory_url = reverse( "browse-directory", url_args={"sha1_git": rev_directory} ) directory_link = None if rev_directory_url is not None: directory_link = gen_link(rev_directory_url, rev_directory) release["directory_link"] = directory_link release["target_link"] = target_link if snapshot_context: snapshot_id = snapshot_context["snapshot_id"] if snapshot_id: - swh_objects.append(SWHObjectInfo(object_type=SNAPSHOT, object_id=snapshot_id)) + swh_objects.append( + SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id) + ) swhids_info = get_swhids_info(swh_objects, snapshot_context) note_header = "None" if len(release_note_lines) > 0: note_header = release_note_lines[0] release["note_header"] = note_header release["note_body"] = "\n".join(release_note_lines[1:]) heading = "Release - %s" % release["name"] if snapshot_context: context_found = "snapshot: %s" % snapshot_context["snapshot_id"] if origin_info: context_found = "origin: %s" % origin_info["url"] heading += " - %s" % context_found return render( request, "browse/release.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"], "swh_object_name": "Release", "swh_object_metadata": release_metadata, "release": release, "snapshot_context": snapshot_context, "show_actions": True, "breadcrumbs": None, "vault_cooking": vault_cooking, "top_right_link": None, "swhids_info": swhids_info, }, ) diff --git a/swh/web/browse/views/revision.py b/swh/web/browse/views/revision.py index fe031404..dd5eb303 100644 --- a/swh/web/browse/views/revision.py +++ b/swh/web/browse/views/revision.py @@ -1,589 +1,586 @@ -# Copyright (C) 2017-2020 The Software Heritage developers +# Copyright (C) 2017-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import hashlib import json import textwrap from django.http import JsonResponse from django.shortcuts import render from django.utils.safestring import mark_safe from swh.model.hashutil import hash_to_bytes -from swh.model.identifiers import ( - CONTENT, - DIRECTORY, - REVISION, - SNAPSHOT, - CoreSWHID, - ObjectType, -) +from swh.model.identifiers import CoreSWHID, ObjectType from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import ( content_display_max_size, format_log_entries, gen_link, gen_person_mail_link, gen_revision_url, get_directory_entries, get_readme_to_display, get_revision_log_url, prepare_content_for_display, request_content, ) from swh.web.common import archive from swh.web.common.exc import NotFoundExc, http_status_code_message from swh.web.common.identifiers import get_swhids_info from swh.web.common.typing import RevisionMetadata, SWHObjectInfo from swh.web.common.utils import ( format_utc_iso_date, gen_path_info, reverse, swh_object_icons, ) def _gen_content_url(revision, query_string, path, snapshot_context): if snapshot_context: query_params = snapshot_context["query_params"] query_params["path"] = path query_params["revision"] = revision["id"] content_url = reverse("browse-origin-content", query_params=query_params) else: content_path = "%s/%s" % (revision["directory"], path) content_url = reverse( "browse-content", url_args={"query_string": query_string}, query_params={"path": content_path}, ) return content_url def _gen_diff_link(idx, diff_anchor, link_text): if idx < _max_displayed_file_diffs: return gen_link(diff_anchor, link_text) else: return link_text # TODO: put in conf _max_displayed_file_diffs = 1000 def _gen_revision_changes_list(revision, changes, snapshot_context): """ Returns a HTML string describing the file changes introduced in a revision. As this string will be displayed in the browse revision view, links to adequate file diffs are also generated. Args: revision (str): hexadecimal representation of a revision identifier changes (list): list of file changes in the revision snapshot_context (dict): optional origin context used to reverse the content urls Returns: A string to insert in a revision HTML view. """ changes_msg = [] for i, change in enumerate(changes): hasher = hashlib.sha1() from_query_string = "" to_query_string = "" diff_id = "diff-" if change["from"]: from_query_string = "sha1_git:" + change["from"]["target"] diff_id += change["from"]["target"] + "-" + change["from_path"] diff_id += "-" if change["to"]: to_query_string = "sha1_git:" + change["to"]["target"] diff_id += change["to"]["target"] + change["to_path"] change["path"] = change["to_path"] or change["from_path"] url_args = { "from_query_string": from_query_string, "to_query_string": to_query_string, } query_params = {"path": change["path"]} change["diff_url"] = reverse( "diff-contents", url_args=url_args, query_params=query_params ) hasher.update(diff_id.encode("utf-8")) diff_id = hasher.hexdigest() change["id"] = diff_id diff_link = "#diff_" + diff_id if change["type"] == "modify": change["content_url"] = _gen_content_url( revision, to_query_string, change["to_path"], snapshot_context ) changes_msg.append( "modified: %s" % _gen_diff_link(i, diff_link, change["to_path"]) ) elif change["type"] == "insert": change["content_url"] = _gen_content_url( revision, to_query_string, change["to_path"], snapshot_context ) changes_msg.append( "new file: %s" % _gen_diff_link(i, diff_link, change["to_path"]) ) elif change["type"] == "delete": parent = archive.lookup_revision(revision["parents"][0]) change["content_url"] = _gen_content_url( parent, from_query_string, change["from_path"], snapshot_context ) changes_msg.append( "deleted: %s" % _gen_diff_link(i, diff_link, change["from_path"]) ) elif change["type"] == "rename": change["content_url"] = _gen_content_url( revision, to_query_string, change["to_path"], snapshot_context ) link_text = change["from_path"] + " → " + change["to_path"] changes_msg.append( "renamed: %s" % _gen_diff_link(i, diff_link, link_text) ) if not changes: changes_msg.append("No changes") return mark_safe("\n".join(changes_msg)) @browse_route( r"revision/(?P[0-9a-f]+)/diff/", view_name="diff-revision", checksum_args=["sha1_git"], ) def _revision_diff(request, sha1_git): """ Browse internal endpoint to compute revision diff """ revision = archive.lookup_revision(sha1_git) snapshot_context = None origin_url = request.GET.get("origin_url", None) if not origin_url: origin_url = request.GET.get("origin", None) timestamp = request.GET.get("timestamp", None) visit_id = request.GET.get("visit_id", None) if origin_url: snapshot_context = get_snapshot_context( origin_url=origin_url, timestamp=timestamp, visit_id=visit_id ) changes = archive.diff_revision(sha1_git) changes_msg = _gen_revision_changes_list(revision, changes, snapshot_context) diff_data = { "total_nb_changes": len(changes), "changes": changes[:_max_displayed_file_diffs], "changes_msg": changes_msg, } return JsonResponse(diff_data) NB_LOG_ENTRIES = 100 @browse_route( r"revision/(?P[0-9a-f]+)/log/", view_name="browse-revision-log", checksum_args=["sha1_git"], ) def revision_log_browse(request, sha1_git): """ Django view that produces an HTML display of the history log for a revision identified by its id. The url that points to it is :http:get:`/browse/revision/(sha1_git)/log/` """ origin_url = request.GET.get("origin_url") snapshot_id = request.GET.get("snapshot") snapshot_context = None if origin_url or snapshot_id: 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 = archive.get_revisions_walker( revs_ordering, sha1_git, max_revs=offset + per_page + 1, state=revs_walker_state, ) rev_log += [rev["id"] for rev in revs_walker] revs_walker_state = revs_walker.export_state() revs = rev_log[offset : offset + per_page] revision_log = archive.lookup_revision_multiple(revs) request.session[session_key] = { "rev_log": rev_log, "revs_walker_state": revs_walker_state, } revs_ordering = request.GET.get("revs_ordering", "") prev_log_url = None if len(rev_log) > offset + per_page: prev_log_url = reverse( "browse-revision-log", url_args={"sha1_git": sha1_git}, query_params={ "per_page": per_page, "offset": offset + per_page, "revs_ordering": revs_ordering or None, }, ) next_log_url = None if offset != 0: next_log_url = reverse( "browse-revision-log", url_args={"sha1_git": sha1_git}, query_params={ "per_page": per_page, "offset": offset - per_page, "revs_ordering": revs_ordering or None, }, ) revision_log_data = format_log_entries(revision_log, per_page) swh_rev_id = str( CoreSWHID(object_type=ObjectType.REVISION, object_id=hash_to_bytes(sha1_git)) ) return render( request, "browse/revision-log.html", { "heading": "Revision history", "swh_object_id": swh_rev_id, "swh_object_name": "Revisions history", "swh_object_metadata": None, "revision_log": revision_log_data, "revs_ordering": revs_ordering, "next_log_url": next_log_url, "prev_log_url": prev_log_url, "breadcrumbs": None, "top_right_link": None, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": True, "swhids_info": None, }, ) @browse_route( r"revision/(?P[0-9a-f]+)/", view_name="browse-revision", checksum_args=["sha1_git"], ) def revision_browse(request, sha1_git): """ Django view that produces an HTML display of a revision identified by its id. The url that points to it is :http:get:`/browse/revision/(sha1_git)/`. """ revision = archive.lookup_revision(sha1_git) origin_info = None snapshot_context = None origin_url = request.GET.get("origin_url") if not origin_url: origin_url = request.GET.get("origin") timestamp = request.GET.get("timestamp") visit_id = request.GET.get("visit_id") snapshot_id = request.GET.get("snapshot_id") if not snapshot_id: snapshot_id = request.GET.get("snapshot") path = request.GET.get("path") dir_id = None dirs, files = [], [] content_data = {} if origin_url: try: snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=visit_id, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=sha1_git, path=path, ) except NotFoundExc as e: raw_rev_url = reverse("browse-revision", url_args={"sha1_git": sha1_git}) error_message = ( "The Software Heritage archive has a revision " "with the hash you provided but the origin " "mentioned in your request appears broken: %s. " "Please check the URL and try again.\n\n" "Nevertheless, you can still browse the revision " "without origin information: %s" % (gen_link(origin_url), gen_link(raw_rev_url)) ) if str(e).startswith("Origin"): raise NotFoundExc(error_message) else: raise e origin_info = snapshot_context["origin_info"] snapshot_id = snapshot_context["snapshot_id"] elif snapshot_id: snapshot_context = get_snapshot_context(snapshot_id) error_info = {"status_code": 200, "description": None} if path: try: file_info = archive.lookup_directory_with_path(revision["directory"], path) if file_info["type"] == "dir": dir_id = file_info["target"] else: query_string = "sha1_git:" + file_info["target"] content_data = request_content(query_string) except NotFoundExc as e: error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" else: dir_id = revision["directory"] if dir_id: path = "" if path is None else (path + "/") dirs, files = get_directory_entries(dir_id) revision_metadata = RevisionMetadata( - object_type=REVISION, + object_type=ObjectType.REVISION, object_id=sha1_git, revision=sha1_git, author=revision["author"]["fullname"] if revision["author"] else "None", author_url=gen_person_mail_link(revision["author"]) if revision["author"] else "None", committer=revision["committer"]["fullname"] if revision["committer"] else "None", committer_url=gen_person_mail_link(revision["committer"]) if revision["committer"] else "None", committer_date=format_utc_iso_date(revision["committer_date"]), date=format_utc_iso_date(revision["date"]), directory=revision["directory"], merge=revision["merge"], metadata=json.dumps( revision["metadata"], sort_keys=True, indent=4, separators=(",", ": ") ), parents=revision["parents"], synthetic=revision["synthetic"], type=revision["type"], snapshot=snapshot_id, origin_url=origin_url, ) message_lines = ["None"] if revision["message"]: message_lines = revision["message"].split("\n") parents = [] for p in revision["parents"]: parent_url = gen_revision_url(p, snapshot_context) parents.append({"id": p, "url": parent_url}) path_info = gen_path_info(path) query_params = snapshot_context["query_params"] if snapshot_context else {} breadcrumbs = [] breadcrumbs.append( { "name": revision["directory"][:7], "url": reverse( "browse-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ), } ) for pi in path_info: query_params["path"] = pi["path"] breadcrumbs.append( { "name": pi["name"], "url": reverse( "browse-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ), } ) vault_cooking = { "directory_context": False, "directory_swhid": None, "revision_context": True, "revision_swhid": f"swh:1:rev:{sha1_git}", } - swh_objects = [SWHObjectInfo(object_type=REVISION, object_id=sha1_git)] + swh_objects = [SWHObjectInfo(object_type=ObjectType.REVISION, object_id=sha1_git)] content = None content_size = None filename = None mimetype = None language = None readme_name = None readme_url = None readme_html = None readmes = {} extra_context = dict(revision_metadata) extra_context["path"] = f"/{path}" if path else None if content_data: breadcrumbs[-1]["url"] = None content_size = content_data["length"] mimetype = content_data["mimetype"] if content_data["raw_data"]: content_display_data = prepare_content_for_display( content_data["raw_data"], content_data["mimetype"], path ) content = content_display_data["content_data"] language = content_display_data["language"] mimetype = content_display_data["mimetype"] if path: filename = path_info[-1]["name"] query_params["filename"] = filename filepath = "/".join(pi["name"] for pi in path_info[:-1]) extra_context["path"] = f"/{filepath}/" if filepath else "/" extra_context["filename"] = filename top_right_link = { "url": reverse( "browse-content-raw", url_args={"query_string": query_string}, query_params={"filename": filename}, ), "icon": swh_object_icons["content"], "text": "Raw File", } swh_objects.append( - SWHObjectInfo(object_type=CONTENT, object_id=file_info["target"]) + SWHObjectInfo(object_type=ObjectType.CONTENT, object_id=file_info["target"]) ) else: for d in dirs: if d["type"] == "rev": d["url"] = reverse( "browse-revision", url_args={"sha1_git": d["target"]} ) else: query_params["path"] = path + d["name"] d["url"] = reverse( "browse-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ) for f in files: query_params["path"] = path + f["name"] f["url"] = reverse( "browse-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ) if f["name"].lower().startswith("readme"): readmes[f["name"]] = f["checksums"]["sha1"] readme_name, readme_url, readme_html = get_readme_to_display(readmes) top_right_link = { "url": get_revision_log_url(sha1_git, snapshot_context), "icon": swh_object_icons["revisions history"], "text": "History", } vault_cooking["directory_context"] = True vault_cooking["directory_swhid"] = f"swh:1:dir:{dir_id}" - swh_objects.append(SWHObjectInfo(object_type=DIRECTORY, object_id=dir_id)) + swh_objects.append( + SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=dir_id) + ) query_params.pop("path", None) diff_revision_url = reverse( "diff-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ) if snapshot_id: - swh_objects.append(SWHObjectInfo(object_type=SNAPSHOT, object_id=snapshot_id)) + swh_objects.append( + SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id) + ) swhids_info = get_swhids_info(swh_objects, snapshot_context, extra_context) heading = "Revision - %s - %s" % ( sha1_git[:7], textwrap.shorten(message_lines[0], width=70), ) if snapshot_context: context_found = "snapshot: %s" % snapshot_context["snapshot_id"] if origin_info: context_found = "origin: %s" % origin_info["url"] heading += " - %s" % context_found return render( request, "browse/revision.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"], "swh_object_name": "Revision", "swh_object_metadata": revision_metadata, "message_header": message_lines[0], "message_body": "\n".join(message_lines[1:]), "parents": parents, "snapshot_context": snapshot_context, "dirs": dirs, "files": files, "content": content, "content_size": content_size, "max_content_size": content_display_max_size, "filename": filename, "encoding": content_data.get("encoding"), "mimetype": mimetype, "language": language, "readme_name": readme_name, "readme_url": readme_url, "readme_html": readme_html, "breadcrumbs": breadcrumbs, "top_right_link": top_right_link, "vault_cooking": vault_cooking, "diff_revision_url": diff_revision_url, "show_actions": True, "swhids_info": swhids_info, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], }, status=error_info["status_code"], ) diff --git a/swh/web/common/archive.py b/swh/web/common/archive.py index 71304a8f..4a090643 100644 --- a/swh/web/common/archive.py +++ b/swh/web/common/archive.py @@ -1,1418 +1,1405 @@ # Copyright (C) 2015-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from collections import defaultdict import itertools import os import re from typing import Any, Dict, Iterable, Iterator, List, Optional, Set, Tuple, Union from urllib.parse import urlparse from swh.model import hashutil -from swh.model.identifiers import ( - CONTENT, - DIRECTORY, - RELEASE, - REVISION, - SNAPSHOT, - CoreSWHID, -) +from swh.model.identifiers import ObjectType, CoreSWHID from swh.model.model import OriginVisit, Revision from swh.storage.algos import diff, revisions_walker from swh.storage.algos.origin import origin_get_latest_visit_status from swh.storage.algos.snapshot import snapshot_get_latest, snapshot_resolve_alias from swh.vault.exc import NotFoundExc as VaultNotFoundExc from swh.web import config from swh.web.common import converters, query -from swh.web.common.exc import BadInputExc, NotFoundExc +from swh.web.common.exc import NotFoundExc from swh.web.common.typing import ( OriginInfo, OriginMetadataInfo, OriginVisitInfo, PagedResult, ) search = config.search() storage = config.storage() vault = config.vault() idx_storage = config.indexer_storage() counters = config.counters() MAX_LIMIT = 50 # Top limit the users can ask for def _first_element(lst): """Returns the first element in the provided list or None if it is empty or None""" return next(iter(lst or []), None) def lookup_multiple_hashes(hashes): """Lookup the passed hashes in a single DB connection, using batch processing. Args: An array of {filename: X, sha1: Y}, string X, hex sha1 string Y. Returns: The same array with elements updated with elem['found'] = true if the hash is present in storage, elem['found'] = false if not. """ hashlist = [hashutil.hash_to_bytes(elem["sha1"]) for elem in hashes] content_missing = storage.content_missing_per_sha1(hashlist) missing = [hashutil.hash_to_hex(x) for x in content_missing] for x in hashes: x.update({"found": True}) for h in hashes: if h["sha1"] in missing: h["found"] = False return hashes def lookup_expression(expression, last_sha1, per_page): """Lookup expression in raw content. Args: expression (str): An expression to lookup through raw indexed content last_sha1 (str): Last sha1 seen per_page (int): Number of results per page Yields: ctags whose content match the expression """ limit = min(per_page, MAX_LIMIT) ctags = idx_storage.content_ctags_search( expression, last_sha1=last_sha1, limit=limit ) for ctag in ctags: ctag = converters.from_swh(ctag, hashess={"id"}) ctag["sha1"] = ctag["id"] ctag.pop("id") yield ctag def lookup_hash(q: str) -> Dict[str, Any]: """Check if the storage contains a given content checksum and return it if found. Args: q: query string of the form Returns: Dict with key found containing the hash info if the hash is present, None if not. """ algo, hash_ = query.parse_hash(q) found = _first_element(storage.content_find({algo: hash_})) if found: content = converters.from_content(found.to_dict()) else: content = None return {"found": content, "algo": algo} def search_hash(q: str) -> Dict[str, bool]: """Search storage for a given content checksum. Args: q: query string of the form Returns: Dict with key found to True or False, according to whether the checksum is present or not """ algo, hash_ = query.parse_hash(q) found = _first_element(storage.content_find({algo: hash_})) return {"found": found is not None} def _lookup_content_sha1(q: str) -> Optional[bytes]: """Given a possible input, query for the content's sha1. Args: q: query string of the form Returns: binary sha1 if found or None """ algo, hash_ = query.parse_hash(q) if algo != "sha1": hashes = _first_element(storage.content_find({algo: hash_})) if not hashes: return None return hashes.sha1 return hash_ def lookup_content_ctags(q): """Return ctags information from a specified content. Args: q: query string of the form Yields: ctags information (dict) list if the content is found. """ sha1 = _lookup_content_sha1(q) if not sha1: return None ctags = list(idx_storage.content_ctags_get([sha1])) if not ctags: return None for ctag in ctags: yield converters.from_swh(ctag, hashess={"id"}) def lookup_content_filetype(q): """Return filetype information from a specified content. Args: q: query string of the form Yields: filetype information (dict) list if the content is found. """ sha1 = _lookup_content_sha1(q) if not sha1: return None filetype = _first_element(list(idx_storage.content_mimetype_get([sha1]))) if not filetype: return None return converters.from_filetype(filetype.to_dict()) def lookup_content_language(q): """Always returns None. This used to return language information from a specified content, but this is currently disabled. Args: q: query string of the form Yields: language information (dict) list if the content is found. """ return None def lookup_content_license(q): """Return license information from a specified content. Args: q: query string of the form Yields: license information (dict) list if the content is found. """ sha1 = _lookup_content_sha1(q) if not sha1: return None licenses = list(idx_storage.content_fossology_license_get([sha1])) if not licenses: return None license_dicts = [license.to_dict() for license in licenses] for license_dict in license_dicts: del license_dict["id"] lic = { "id": sha1, "facts": license_dicts, } return converters.from_swh(lic, hashess={"id"}) def lookup_origin(origin: OriginInfo) -> OriginInfo: """Return information about the origin matching dict origin. Args: origin: origin's dict with 'url' key Returns: origin information as dict. """ origin_urls = [origin["url"]] if origin["url"]: # handle case when user provided an origin url with a trailing # slash while the url in storage does not have it (e.g. GitHub) if origin["url"].endswith("/"): origin_urls.append(origin["url"][:-1]) # handle case when user provided an origin url without a trailing # slash while the url in storage have it (e.g. Debian source package) else: origin_urls.append(f"{origin['url']}/") try: # handle case where the "://" character sequence was mangled into ":/" parsed_url = urlparse(origin["url"]) if ( parsed_url.scheme and not parsed_url.netloc and origin["url"].startswith(f"{parsed_url.scheme}:/") and not origin["url"].startswith(f"{parsed_url.scheme}://") ): origin_urls.append( origin["url"].replace( f"{parsed_url.scheme}:/", f"{parsed_url.scheme}://" ) ) except Exception: pass origins = [o for o in storage.origin_get(origin_urls) if o is not None] if not origins: msg = "Origin with url %s not found!" % origin["url"] raise NotFoundExc(msg) return converters.from_origin(origins[0].to_dict()) def lookup_origins( page_token: Optional[str], limit: int = 100 ) -> PagedResult[OriginInfo]: """Get list of archived software origins in a paginated way. Origins are sorted by id before returning them Args: origin_from (int): The minimum id of the origins to return origin_count (int): The maximum number of origins to return Returns: Page of OriginInfo """ page = storage.origin_list(page_token=page_token, limit=limit) return PagedResult( [converters.from_origin(o.to_dict()) for o in page.results], next_page_token=page.next_page_token, ) def search_origin( url_pattern: str, use_ql: bool = False, limit: int = 50, with_visit: bool = False, visit_types: Optional[List[str]] = None, page_token: Optional[str] = None, ) -> Tuple[List[OriginInfo], Optional[str]]: """Search for origins whose urls contain a provided string pattern or match a provided regular expression. Args: url_pattern: the string pattern to search for in origin urls use_ql: whether to use swh search query language or not limit: the maximum number of found origins to return with_visit: Whether origins with no visit are to be filtered out visit_types: Only origins having any of the provided visit types (e.g. git, svn, pypi) will be returned page_token: opaque string used to get the next results of a search Returns: list of origin information as dict. """ if page_token: assert isinstance(page_token, str) if search: if config.get_config()["search_config"].get("enable_ql") and use_ql: page_result = search.origin_search( query=url_pattern, page_token=page_token, with_visit=with_visit, visit_types=visit_types, limit=limit, ) else: page_result = search.origin_search( url_pattern=url_pattern, page_token=page_token, with_visit=with_visit, visit_types=visit_types, limit=limit, ) origins = [converters.from_origin(ori_dict) for ori_dict in page_result.results] else: # Fallback to swh-storage if swh-search is not configured search_words = [re.escape(word) for word in url_pattern.split()] if len(search_words) >= 7: url_pattern = ".*".join(search_words) else: pattern_parts = [] for permut in itertools.permutations(search_words): pattern_parts.append(".*".join(permut)) url_pattern = "|".join(pattern_parts) page_result = storage.origin_search( url_pattern, page_token=page_token, with_visit=with_visit, limit=limit, visit_types=visit_types, regexp=True, ) origins = [converters.from_origin(ori.to_dict()) for ori in page_result.results] return (origins, page_result.next_page_token) def search_origin_metadata( fulltext: str, limit: int = 50 ) -> Iterable[OriginMetadataInfo]: """Search for origins whose metadata match a provided string pattern. Args: fulltext: the string pattern to search for in origin metadata limit: the maximum number of found origins to return Returns: Iterable of origin metadata information for existing origins """ results = [] if ( search and config.get_config()["search_config"]["metadata_backend"] == "swh-search" ): page_result = search.origin_search(metadata_pattern=fulltext, limit=limit,) matches = idx_storage.origin_intrinsic_metadata_get( [r["url"] for r in page_result.results] ) else: matches = idx_storage.origin_intrinsic_metadata_search_fulltext( conjunction=[fulltext], limit=limit ) matches = [match.to_dict() for match in matches] origins = storage.origin_get([match["id"] for match in matches]) for origin, match in zip(origins, matches): if not origin: continue match["from_revision"] = hashutil.hash_to_hex(match["from_revision"]) del match["id"] results.append(OriginMetadataInfo(url=origin.url, metadata=match)) return results def lookup_origin_intrinsic_metadata(origin_url: str) -> Dict[str, Any]: """Return intrinsic metadata for origin whose origin matches given origin. Args: origin_url: origin url Raises: NotFoundExc when the origin is not found Returns: origin metadata. """ origins = [origin_url] origin_info = storage.origin_get(origins)[0] if not origin_info: raise NotFoundExc(f"Origin with url {origin_url} not found!") match = _first_element(idx_storage.origin_intrinsic_metadata_get(origins)) result = {} if match: result = match.metadata return result def _to_sha1_bin(sha1_hex): _, sha1_git_bin = query.parse_hash_with_algorithms_or_throws( sha1_hex, ["sha1"], "Only sha1_git is supported." # HACK: sha1_git really ) return sha1_git_bin def _check_directory_exists(sha1_git, sha1_git_bin): if len(list(storage.directory_missing([sha1_git_bin]))): raise NotFoundExc("Directory with sha1_git %s not found" % sha1_git) def lookup_directory(sha1_git): """Return information about the directory with id sha1_git. Args: sha1_git as string Returns: directory information as dict. """ empty_dir_sha1 = "4b825dc642cb6eb9a060e54bf8d69288fbee4904" if sha1_git == empty_dir_sha1: return [] sha1_git_bin = _to_sha1_bin(sha1_git) _check_directory_exists(sha1_git, sha1_git_bin) directory_entries = storage.directory_ls(sha1_git_bin) return map(converters.from_directory_entry, directory_entries) def lookup_directory_with_path(sha1_git: str, path: str) -> Dict[str, Any]: """Return directory information for entry with specified path w.r.t. root directory pointed by sha1_git Args: sha1_git: sha1_git corresponding to the directory to which we append paths to (hopefully) find the entry path: the relative path to the entry starting from the root directory pointed by sha1_git Returns: Directory entry information as dict. Raises: NotFoundExc if the directory entry is not found """ sha1_git_bin = _to_sha1_bin(sha1_git) _check_directory_exists(sha1_git, sha1_git_bin) paths = path.strip(os.path.sep).split(os.path.sep) queried_dir = storage.directory_entry_get_by_path( sha1_git_bin, [p.encode("utf-8") for p in paths] ) if not queried_dir: raise NotFoundExc( f"Directory entry with path {path} from root directory {sha1_git} not found" ) return converters.from_directory_entry(queried_dir) def lookup_release(release_sha1_git: str) -> Dict[str, Any]: """Return information about the release with sha1 release_sha1_git. Args: release_sha1_git: The release's sha1 as hexadecimal Returns: Release information as dict. Raises: ValueError if the identifier provided is not of sha1 nature. """ sha1_git_bin = _to_sha1_bin(release_sha1_git) release = _first_element(storage.release_get([sha1_git_bin])) if not release: raise NotFoundExc(f"Release with sha1_git {release_sha1_git} not found.") return converters.from_release(release) def lookup_release_multiple(sha1_git_list) -> Iterator[Optional[Dict[str, Any]]]: """Return information about the releases identified with their sha1_git identifiers. Args: sha1_git_list: A list of release sha1_git identifiers Returns: Iterator of Release metadata information as dict. Raises: ValueError if the identifier provided is not of sha1 nature. """ sha1_bin_list = [_to_sha1_bin(sha1_git) for sha1_git in sha1_git_list] releases = storage.release_get(sha1_bin_list) for r in releases: if r is not None: yield converters.from_release(r) else: yield None def lookup_revision(rev_sha1_git) -> Dict[str, Any]: """Return information about the revision with sha1 revision_sha1_git. Args: revision_sha1_git: The revision's sha1 as hexadecimal Returns: Revision information as dict. Raises: ValueError if the identifier provided is not of sha1 nature. NotFoundExc if there is no revision with the provided sha1_git. """ sha1_git_bin = _to_sha1_bin(rev_sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision with sha1_git {rev_sha1_git} not found.") return converters.from_revision(revision) def lookup_revision_multiple(sha1_git_list) -> Iterator[Optional[Dict[str, Any]]]: """Return information about the revisions identified with their sha1_git identifiers. Args: sha1_git_list: A list of revision sha1_git identifiers Yields: revision information as dict if the revision exists, None otherwise. Raises: ValueError if the identifier provided is not of sha1 nature. """ sha1_bin_list = [_to_sha1_bin(sha1_git) for sha1_git in sha1_git_list] revisions = storage.revision_get(sha1_bin_list) for revision in revisions: if revision is not None: yield converters.from_revision(revision) else: yield None def lookup_revision_message(rev_sha1_git) -> Dict[str, bytes]: """Return the raw message of the revision with sha1 revision_sha1_git. Args: revision_sha1_git: The revision's sha1 as hexadecimal Returns: Decoded revision message as dict {'message': } Raises: ValueError if the identifier provided is not of sha1 nature. NotFoundExc if the revision is not found, or if it has no message """ sha1_git_bin = _to_sha1_bin(rev_sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision with sha1_git {rev_sha1_git} not found.") if not revision.message: raise NotFoundExc(f"No message for revision with sha1_git {rev_sha1_git}.") return {"message": revision.message} def _lookup_revision_id_by(origin, branch_name, timestamp): def _get_snapshot_branch(snapshot, branch_name): snapshot = lookup_snapshot( visit["snapshot"], branches_from=branch_name, branches_count=10, branch_name_exclude_prefix=None, ) branch = None if branch_name in snapshot["branches"]: branch = snapshot["branches"][branch_name] return branch if isinstance(origin, int): origin = {"id": origin} elif isinstance(origin, str): origin = {"url": origin} else: raise TypeError('"origin" must be an int or a string.') from swh.web.common.origin_visits import get_origin_visit visit = get_origin_visit(origin, visit_ts=timestamp) branch = _get_snapshot_branch(visit["snapshot"], branch_name) rev_id = None if branch and branch["target_type"] == "revision": rev_id = branch["target"] elif branch and branch["target_type"] == "alias": branch = _get_snapshot_branch(visit["snapshot"], branch["target"]) if branch and branch["target_type"] == "revision": rev_id = branch["target"] if not rev_id: raise NotFoundExc( "Revision for origin %s and branch %s not found." % (origin.get("url"), branch_name) ) return rev_id def lookup_revision_by(origin, branch_name="HEAD", timestamp=None): """Lookup revision by origin, snapshot branch name and visit timestamp. If branch_name is not provided, lookup using 'HEAD' as default. If timestamp is not provided, use the most recent. Args: origin (Union[int,str]): origin of the revision branch_name (str): snapshot branch name timestamp (str/int): origin visit time frame Returns: dict: The revision matching the criterions Raises: NotFoundExc if no revision corresponds to the criterion """ rev_id = _lookup_revision_id_by(origin, branch_name, timestamp) return lookup_revision(rev_id) def lookup_revision_log(rev_sha1_git, limit): """Lookup revision log by revision id. Args: rev_sha1_git (str): The revision's sha1 as hexadecimal limit (int): the maximum number of revisions returned Returns: list: Revision log as list of revision dicts Raises: ValueError: if the identifier provided is not of sha1 nature. swh.web.common.exc.NotFoundExc: if there is no revision with the provided sha1_git. """ lookup_revision(rev_sha1_git) sha1_git_bin = _to_sha1_bin(rev_sha1_git) revision_entries = storage.revision_log([sha1_git_bin], limit) return map(converters.from_revision, revision_entries) def lookup_revision_log_by(origin, branch_name, timestamp, limit): """Lookup revision by origin, snapshot branch name and visit timestamp. Args: origin (Union[int,str]): origin of the revision branch_name (str): snapshot branch timestamp (str/int): origin visit time frame limit (int): the maximum number of revisions returned Returns: list: Revision log as list of revision dicts Raises: swh.web.common.exc.NotFoundExc: if no revision corresponds to the criterion """ rev_id = _lookup_revision_id_by(origin, branch_name, timestamp) return lookup_revision_log(rev_id, limit) def lookup_revision_with_context_by( origin, branch_name, timestamp, sha1_git, limit=100 ): """Return information about revision sha1_git, limited to the sub-graph of all transitive parents of sha1_git_root. sha1_git_root being resolved through the lookup of a revision by origin, branch_name and ts. In other words, sha1_git is an ancestor of sha1_git_root. Args: - origin: origin of the revision. - branch_name: revision's branch. - timestamp: revision's time frame. - sha1_git: one of sha1_git_root's ancestors. - limit: limit the lookup to 100 revisions back. Returns: Pair of (root_revision, revision). Information on sha1_git if it is an ancestor of sha1_git_root including children leading to sha1_git_root Raises: - BadInputExc in case of unknown algo_hash or bad hash. - NotFoundExc if either revision is not found or if sha1_git is not an ancestor of sha1_git_root. """ rev_root_id = _lookup_revision_id_by(origin, branch_name, timestamp) rev_root_id_bin = hashutil.hash_to_bytes(rev_root_id) rev_root = storage.revision_get([rev_root_id_bin])[0] return ( converters.from_revision(rev_root) if rev_root else None, lookup_revision_with_context(rev_root, sha1_git, limit), ) def lookup_revision_with_context( sha1_git_root: Union[str, Dict[str, Any], Revision], sha1_git: str, limit: int = 100 ) -> Dict[str, Any]: """Return information about revision sha1_git, limited to the sub-graph of all transitive parents of sha1_git_root. In other words, sha1_git is an ancestor of sha1_git_root. Args: sha1_git_root: latest revision. The type is either a sha1 (as an hex string) or a non converted dict. sha1_git: one of sha1_git_root's ancestors limit: limit the lookup to 100 revisions back Returns: Information on sha1_git if it is an ancestor of sha1_git_root including children leading to sha1_git_root Raises: BadInputExc in case of unknown algo_hash or bad hash NotFoundExc if either revision is not found or if sha1_git is not an ancestor of sha1_git_root """ sha1_git_bin = _to_sha1_bin(sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision {sha1_git} not found") if isinstance(sha1_git_root, str): sha1_git_root_bin = _to_sha1_bin(sha1_git_root) revision_root = storage.revision_get([sha1_git_root_bin])[0] if not revision_root: raise NotFoundExc(f"Revision root {sha1_git_root} not found") elif isinstance(sha1_git_root, Revision): sha1_git_root_bin = sha1_git_root.id else: sha1_git_root_bin = sha1_git_root["id"] revision_log = storage.revision_log([sha1_git_root_bin], limit) parents: Dict[str, List[str]] = {} children = defaultdict(list) for rev in revision_log: rev_id = rev["id"] parents[rev_id] = [] for parent_id in rev["parents"]: parents[rev_id].append(parent_id) children[parent_id].append(rev_id) if revision.id not in parents: raise NotFoundExc(f"Revision {sha1_git} is not an ancestor of {sha1_git_root}") revision_d = revision.to_dict() revision_d["children"] = children[revision.id] return converters.from_revision(revision_d) def lookup_directory_with_revision(sha1_git, dir_path=None, with_data=False): """Return information on directory pointed by revision with sha1_git. If dir_path is not provided, display top level directory. Otherwise, display the directory pointed by dir_path (if it exists). Args: sha1_git: revision's hash. dir_path: optional directory pointed to by that revision. with_data: boolean that indicates to retrieve the raw data if the path resolves to a content. Default to False (for the api) Returns: Information on the directory pointed to by that revision. Raises: BadInputExc in case of unknown algo_hash or bad hash. NotFoundExc either if the revision is not found or the path referenced does not exist. NotImplementedError in case of dir_path exists but do not reference a type 'dir' or 'file'. """ sha1_git_bin = _to_sha1_bin(sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision {sha1_git} not found") dir_sha1_git_bin = revision.directory if dir_path: paths = dir_path.strip(os.path.sep).split(os.path.sep) entity = storage.directory_entry_get_by_path( dir_sha1_git_bin, list(map(lambda p: p.encode("utf-8"), paths)) ) if not entity: raise NotFoundExc( "Directory or File '%s' pointed to by revision %s not found" % (dir_path, sha1_git) ) else: entity = {"type": "dir", "target": dir_sha1_git_bin} if entity["type"] == "dir": directory_entries = storage.directory_ls(entity["target"]) or [] return { "type": "dir", "path": "." if not dir_path else dir_path, "revision": sha1_git, "content": list(map(converters.from_directory_entry, directory_entries)), } elif entity["type"] == "file": # content content = _first_element(storage.content_find({"sha1_git": entity["target"]})) if not content: raise NotFoundExc(f"Content not found for revision {sha1_git}") content_d = content.to_dict() if with_data: data = storage.content_get_data(content.sha1) if data: content_d["data"] = data return { "type": "file", "path": "." if not dir_path else dir_path, "revision": sha1_git, "content": converters.from_content(content_d), } elif entity["type"] == "rev": # revision revision = storage.revision_get([entity["target"]])[0] return { "type": "rev", "path": "." if not dir_path else dir_path, "revision": sha1_git, "content": converters.from_revision(revision) if revision else None, } else: raise NotImplementedError("Entity of type %s not implemented." % entity["type"]) def lookup_content(q: str) -> Dict[str, Any]: """Lookup the content designed by q. Args: q: The release's sha1 as hexadecimal Raises: NotFoundExc if the requested content is not found """ algo, hash_ = query.parse_hash(q) c = _first_element(storage.content_find({algo: hash_})) if not c: hhex = hashutil.hash_to_hex(hash_) raise NotFoundExc(f"Content with {algo} checksum equals to {hhex} not found!") return converters.from_content(c.to_dict()) def lookup_content_raw(q: str) -> Dict[str, Any]: """Lookup the content defined by q. Args: q: query string of the form Returns: dict with 'sha1' and 'data' keys. data representing its raw data decoded. Raises: NotFoundExc if the requested content is not found or if the content bytes are not available in the storage """ c = lookup_content(q) content_sha1_bytes = hashutil.hash_to_bytes(c["checksums"]["sha1"]) content_data = storage.content_get_data(content_sha1_bytes) if content_data is None: algo, hash_ = query.parse_hash(q) raise NotFoundExc( f"Bytes of content with {algo} checksum equals " f"to {hashutil.hash_to_hex(hash_)} are not available!" ) return converters.from_content({"sha1": content_sha1_bytes, "data": content_data}) def stat_counters(): """Return the stat counters for Software Heritage Returns: A dict mapping textual labels to integer values. """ res = {} if counters and config.get_config()["counters_backend"] == "swh-counters": res = counters.get_counts( ["origin", "revision", "content", "directory", "release", "person"] ) else: res = storage.stat_counters() return res def _lookup_origin_visits( origin_url: str, last_visit: Optional[int] = None, limit: int = 10 ) -> Iterator[OriginVisit]: """Yields the origin origins' visits. Args: origin_url (str): origin to list visits for last_visit (int): last visit to lookup from limit (int): Number of elements max to display Yields: OriginVisit for that origin """ limit = min(limit, MAX_LIMIT) page_token: Optional[str] if last_visit is not None: page_token = str(last_visit) else: page_token = None visit_page = storage.origin_visit_get( origin_url, page_token=page_token, limit=limit ) yield from visit_page.results def lookup_origin_visits( origin: str, last_visit: Optional[int] = None, per_page: int = 10 ) -> Iterator[OriginVisitInfo]: """Yields the origin origins' visits. Args: origin: origin to list visits for Yields: Dictionaries of origin_visit for that origin """ for visit in _lookup_origin_visits(origin, last_visit=last_visit, limit=per_page): visit_status = storage.origin_visit_status_get_latest(origin, visit.visit) yield converters.from_origin_visit( {**visit_status.to_dict(), "type": visit.type} ) def lookup_origin_visit_latest( origin_url: str, require_snapshot: bool = False, type: Optional[str] = None, allowed_statuses: Optional[List[str]] = None, ) -> Optional[OriginVisitInfo]: """Return the origin's latest visit Args: origin_url: origin to list visits for type: Optional visit type to filter on (e.g git, tar, dsc, svn, hg, npm, pypi, ...) allowed_statuses: list of visit statuses considered to find the latest visit. For instance, ``allowed_statuses=['full']`` will only consider visits that have successfully run to completion. require_snapshot: filter out origins without a snapshot Returns: The origin visit info as dict if found """ visit_status = origin_get_latest_visit_status( storage, origin_url, type=type, allowed_statuses=allowed_statuses, require_snapshot=require_snapshot, ) return ( converters.from_origin_visit(visit_status.to_dict()) if visit_status else None ) def lookup_origin_visit(origin_url: str, visit_id: int) -> OriginVisitInfo: """Return information about visit visit_id with origin origin. Args: origin: origin concerned by the visit visit_id: the visit identifier to lookup Yields: The dict origin_visit concerned """ visit = storage.origin_visit_get_by(origin_url, visit_id) visit_status = storage.origin_visit_status_get_latest(origin_url, visit_id) if not visit: raise NotFoundExc( f"Origin {origin_url} or its visit with id {visit_id} not found!" ) return converters.from_origin_visit({**visit_status.to_dict(), "type": visit.type}) def lookup_snapshot_sizes( snapshot_id: str, branch_name_exclude_prefix: Optional[str] = "refs/pull/" ) -> Dict[str, int]: """Count the number of branches in the snapshot with the given id Args: snapshot_id (str): sha1 identifier of the snapshot Returns: dict: A dict whose keys are the target types of branches and values their corresponding amount """ snapshot_id_bin = _to_sha1_bin(snapshot_id) snapshot_sizes = dict.fromkeys(("alias", "release", "revision"), 0) branch_counts = storage.snapshot_count_branches( snapshot_id_bin, branch_name_exclude_prefix.encode() if branch_name_exclude_prefix else None, ) # remove possible None key returned by snapshot_count_branches # when null branches are present in the snapshot branch_counts.pop(None, None) snapshot_sizes.update(branch_counts) return snapshot_sizes def lookup_snapshot( snapshot_id: str, branches_from: str = "", branches_count: int = 1000, target_types: Optional[List[str]] = None, branch_name_include_substring: Optional[str] = None, branch_name_exclude_prefix: Optional[str] = "refs/pull/", ) -> Dict[str, Any]: """Return information about a snapshot, aka the list of named branches found during a specific visit of an origin. Args: snapshot_id: sha1 identifier of the snapshot branches_from: optional parameter used to skip branches whose name is lesser than it before returning them branches_count: optional parameter used to restrain the amount of returned branches target_types: optional parameter used to filter the target types of branch to return (possible values that can be contained in that list are `'content', 'directory', 'revision', 'release', 'snapshot', 'alias'`) branch_name_include_substring: if provided, only return branches whose name contains given substring branch_name_exclude_prefix: if provided, do not return branches whose name starts with given pattern Raises: NotFoundExc if the given snapshot_id is missing Returns: A dict filled with the snapshot content. """ snapshot_id_bin = _to_sha1_bin(snapshot_id) if storage.snapshot_missing([snapshot_id_bin]): raise NotFoundExc(f"Snapshot with id {snapshot_id} not found!") partial_branches = storage.snapshot_get_branches( snapshot_id_bin, branches_from.encode(), branches_count, target_types, branch_name_include_substring.encode() if branch_name_include_substring else None, branch_name_exclude_prefix.encode() if branch_name_exclude_prefix else None, ) return ( converters.from_partial_branches(partial_branches) if partial_branches else None ) def lookup_latest_origin_snapshot( origin: str, allowed_statuses: List[str] = None ) -> Optional[Dict[str, Any]]: """Return information about the latest snapshot of an origin. .. warning:: At most 1000 branches contained in the snapshot will be returned for performance reasons. Args: origin: URL or integer identifier of the origin allowed_statuses: list of visit statuses considered to find the latest snapshot for the visit. For instance, ``allowed_statuses=['full']`` will only consider visits that have successfully run to completion. Returns: A dict filled with the snapshot content. """ snp = snapshot_get_latest( storage, origin, allowed_statuses=allowed_statuses, branches_count=1000 ) return converters.from_snapshot(snp.to_dict()) if snp is not None else None def lookup_snapshot_alias( snapshot_id: str, alias_name: str ) -> Optional[Dict[str, Any]]: """Try to resolve a branch alias in a snapshot. Args: snapshot_id: hexadecimal representation of a snapshot id alias_name: name of the branch alias to resolve Returns: Target branch information or None if the alias does not exist or target a dangling branch. """ resolved_alias = snapshot_resolve_alias( storage, _to_sha1_bin(snapshot_id), alias_name.encode() ) return ( converters.from_swh(resolved_alias.to_dict(), hashess={"target"}) if resolved_alias is not None else None ) def lookup_revision_through(revision, limit=100): """Retrieve a revision from the criterion stored in revision dictionary. Args: revision: Dictionary of criterion to lookup the revision with. Here are the supported combination of possible values: - origin_url, branch_name, ts, sha1_git - origin_url, branch_name, ts - sha1_git_root, sha1_git - sha1_git Returns: None if the revision is not found or the actual revision. """ if ( "origin_url" in revision and "branch_name" in revision and "ts" in revision and "sha1_git" in revision ): return lookup_revision_with_context_by( revision["origin_url"], revision["branch_name"], revision["ts"], revision["sha1_git"], limit, ) if "origin_url" in revision and "branch_name" in revision and "ts" in revision: return lookup_revision_by( revision["origin_url"], revision["branch_name"], revision["ts"] ) if "sha1_git_root" in revision and "sha1_git" in revision: return lookup_revision_with_context( revision["sha1_git_root"], revision["sha1_git"], limit ) if "sha1_git" in revision: return lookup_revision(revision["sha1_git"]) # this should not happen raise NotImplementedError("Should not happen!") def lookup_directory_through_revision(revision, path=None, limit=100, with_data=False): """Retrieve the directory information from the revision. Args: revision: dictionary of criterion representing a revision to lookup path: directory's path to lookup. limit: optional query parameter to limit the revisions log (default to 100). For now, note that this limit could impede the transitivity conclusion about sha1_git not being an ancestor of. with_data: indicate to retrieve the content's raw data if path resolves to a content. Returns: The directory pointing to by the revision criterions at path. """ rev = lookup_revision_through(revision, limit) if not rev: raise NotFoundExc("Revision with criterion %s not found!" % revision) return (rev["id"], lookup_directory_with_revision(rev["id"], path, with_data)) def _vault_request(vault_fn, bundle_type: str, swhid: CoreSWHID, **kwargs): try: return vault_fn(bundle_type, swhid, **kwargs) except VaultNotFoundExc: return None def vault_cook(bundle_type: str, swhid: CoreSWHID, email=None): """Cook a vault bundle. """ return _vault_request(vault.cook, bundle_type, swhid, email=email) def vault_fetch(bundle_type: str, swhid: CoreSWHID): """Fetch a vault bundle. """ return _vault_request(vault.fetch, bundle_type, swhid) def vault_progress(bundle_type: str, swhid: CoreSWHID): """Get the current progress of a vault bundle. """ return _vault_request(vault.progress, bundle_type, swhid) def diff_revision(rev_id): """Get the list of file changes (insertion / deletion / modification / renaming) for a particular revision. """ rev_sha1_git_bin = _to_sha1_bin(rev_id) changes = diff.diff_revision(storage, rev_sha1_git_bin, track_renaming=True) for change in changes: change["from"] = converters.from_directory_entry(change["from"]) change["to"] = converters.from_directory_entry(change["to"]) if change["from_path"]: change["from_path"] = change["from_path"].decode("utf-8") if change["to_path"]: change["to_path"] = change["to_path"].decode("utf-8") return changes class _RevisionsWalkerProxy(object): """ Proxy class wrapping a revisions walker iterator from swh-storage and performing needed conversions. """ def __init__(self, rev_walker_type, rev_start, *args, **kwargs): rev_start_bin = hashutil.hash_to_bytes(rev_start) self.revisions_walker = revisions_walker.get_revisions_walker( rev_walker_type, storage, rev_start_bin, *args, **kwargs ) def export_state(self): return self.revisions_walker.export_state() def __next__(self): return converters.from_revision(next(self.revisions_walker)) def __iter__(self): return self def get_revisions_walker(rev_walker_type, rev_start, *args, **kwargs): """ Utility function to instantiate a revisions walker of a given type, see :mod:`swh.storage.algos.revisions_walker`. Args: rev_walker_type (str): the type of revisions walker to return, possible values are: ``committer_date``, ``dfs``, ``dfs_post``, ``bfs`` and ``path`` rev_start (str): hexadecimal representation of a revision identifier args (list): position arguments to pass to the revisions walker constructor kwargs (dict): keyword arguments to pass to the revisions walker constructor """ # first check if the provided revision is valid lookup_revision(rev_start) return _RevisionsWalkerProxy(rev_walker_type, rev_start, *args, **kwargs) -def lookup_object(object_type: str, object_id: str) -> Dict[str, Any]: +def lookup_object(object_type: ObjectType, object_id: str) -> Dict[str, Any]: """ Utility function for looking up an object in the archive by its type and id. Args: object_type (str): the type of object to lookup, either *content*, *directory*, *release*, *revision* or *snapshot* object_id (str): the *sha1_git* checksum identifier in hexadecimal form of the object to lookup Returns: Dict[str, Any]: A dictionary describing the object or a list of dictionary for the directory object type. Raises: swh.web.common.exc.NotFoundExc: if the object could not be found in the archive BadInputExc: if the object identifier is invalid """ - if object_type == CONTENT: + if object_type == ObjectType.CONTENT: return lookup_content(f"sha1_git:{object_id}") - elif object_type == DIRECTORY: + elif object_type == ObjectType.DIRECTORY: return {"id": object_id, "content": list(lookup_directory(object_id))} - elif object_type == RELEASE: + elif object_type == ObjectType.RELEASE: return lookup_release(object_id) - elif object_type == REVISION: + elif object_type == ObjectType.REVISION: return lookup_revision(object_id) - elif object_type == SNAPSHOT: + elif object_type == ObjectType.SNAPSHOT: return lookup_snapshot(object_id) - - raise BadInputExc( - ( - "Invalid swh object type! Valid types are " - f"{CONTENT}, {DIRECTORY}, {RELEASE} " - f"{REVISION} or {SNAPSHOT}." - ) - ) + else: + raise ValueError(f"Unexpected object type variant: {object_type}") def lookup_missing_hashes(grouped_swhids: Dict[str, List[bytes]]) -> Set[str]: """Lookup missing Software Heritage persistent identifier hash, using batch processing. Args: A dictionary with: keys: object types values: object hashes Returns: A set(hexadecimal) of the hashes not found in the storage """ missing_hashes = [] for obj_type, obj_ids in grouped_swhids.items(): - if obj_type == CONTENT: + if obj_type == ObjectType.CONTENT: missing_hashes.append(storage.content_missing_per_sha1_git(obj_ids)) - elif obj_type == DIRECTORY: + elif obj_type == ObjectType.DIRECTORY: missing_hashes.append(storage.directory_missing(obj_ids)) - elif obj_type == REVISION: + elif obj_type == ObjectType.REVISION: missing_hashes.append(storage.revision_missing(obj_ids)) - elif obj_type == RELEASE: + elif obj_type == ObjectType.RELEASE: missing_hashes.append(storage.release_missing(obj_ids)) - elif obj_type == SNAPSHOT: + elif obj_type == ObjectType.SNAPSHOT: missing_hashes.append(storage.snapshot_missing(obj_ids)) missing = set( map(lambda x: hashutil.hash_to_hex(x), itertools.chain(*missing_hashes)) ) return missing def lookup_origins_by_sha1s(sha1s: List[str]) -> Iterator[Optional[OriginInfo]]: """Lookup origins from the sha1 hash values of their URLs. Args: sha1s: list of sha1s hexadecimal representation Yields: origin information as dict """ sha1s_bytes = [hashutil.hash_to_bytes(sha1) for sha1 in sha1s] origins = storage.origin_get_by_sha1(sha1s_bytes) for origin in origins: yield converters.from_origin(origin) diff --git a/swh/web/common/converters.py b/swh/web/common/converters.py index 3b1dcfb7..21c15b1c 100644 --- a/swh/web/common/converters.py +++ b/swh/web/common/converters.py @@ -1,395 +1,409 @@ -# Copyright (C) 2015-2020 The Software Heritage developers +# Copyright (C) 2015-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import datetime import json from typing import Any, Dict, Union +from django.core.serializers.json import DjangoJSONEncoder + from swh.core.utils import decode_with_escape from swh.model import hashutil +from swh.model.identifiers import ObjectType from swh.model.model import RawExtrinsicMetadata, Release, Revision from swh.storage.interface import PartialBranches from swh.web.common.typing import OriginInfo, OriginVisitInfo def _group_checksums(data): """Groups checksums values computed from hash functions used in swh and stored in data dict under a single entry 'checksums' """ if data: checksums = {} for hash in hashutil.ALGORITHMS: if hash in data and data[hash]: checksums[hash] = data[hash] del data[hash] if len(checksums) > 0: data["checksums"] = checksums def fmap(f, data): """Map f to data at each level. This must keep the origin data structure type: - map -> map - dict -> dict - list -> list - None -> None Args: f: function that expects one argument. data: data to traverse to apply the f function. list, map, dict or bare value. Returns: The same data-structure with modified values by the f function. """ if data is None: return data if isinstance(data, map): return map(lambda y: fmap(f, y), (x for x in data)) if isinstance(data, list): return [fmap(f, x) for x in data] if isinstance(data, tuple): return tuple(fmap(f, x) for x in data) if isinstance(data, dict): return {k: fmap(f, v) for (k, v) in data.items()} return f(data) def from_swh( dict_swh, hashess={}, bytess={}, dates={}, blacklist={}, removables_if_empty={}, empty_dict={}, empty_list={}, convert={}, convert_fn=lambda x: x, ): """Convert from a swh dictionary to something reasonably json serializable. Args: dict_swh: the origin dictionary needed to be transformed hashess: list/set of keys representing hashes values (sha1, sha256, sha1_git, etc...) as bytes. Those need to be transformed in hexadecimal string bytess: list/set of keys representing bytes values which needs to be decoded blacklist: set of keys to filter out from the conversion convert: set of keys whose associated values need to be converted using convert_fn convert_fn: the conversion function to apply on the value of key in 'convert' The remaining keys are copied as is in the output. Returns: dictionary equivalent as dict_swh only with its keys converted. """ def convert_hashes_bytes(v): """v is supposedly a hash as bytes, returns it converted in hex. """ if isinstance(v, bytes): return hashutil.hash_to_hex(v) return v def convert_bytes(v): """v is supposedly a bytes string, decode as utf-8. FIXME: Improve decoding policy. If not utf-8, break! """ if isinstance(v, bytes): return v.decode("utf-8") return v def convert_date(v): """ Args: v (dict or datatime): either: - a dict with three keys: - timestamp (dict or integer timestamp) - offset - negative_utc - or, a datetime We convert it to a human-readable string """ if not v: return v if isinstance(v, datetime.datetime): return v.isoformat() tz = datetime.timezone(datetime.timedelta(minutes=v["offset"])) swh_timestamp = v["timestamp"] if isinstance(swh_timestamp, dict): date = datetime.datetime.fromtimestamp(swh_timestamp["seconds"], tz=tz) else: date = datetime.datetime.fromtimestamp(swh_timestamp, tz=tz) datestr = date.isoformat() if v["offset"] == 0 and v["negative_utc"]: # remove the rightmost + and replace it with a - return "-".join(datestr.rsplit("+", 1)) return datestr if not dict_swh: return dict_swh new_dict = {} for key, value in dict_swh.items(): if key in blacklist or (key in removables_if_empty and not value): continue if key in dates: new_dict[key] = convert_date(value) elif key in convert: new_dict[key] = convert_fn(value) elif isinstance(value, dict): new_dict[key] = from_swh( value, hashess=hashess, bytess=bytess, dates=dates, blacklist=blacklist, removables_if_empty=removables_if_empty, empty_dict=empty_dict, empty_list=empty_list, convert=convert, convert_fn=convert_fn, ) elif key in hashess: new_dict[key] = fmap(convert_hashes_bytes, value) elif key in bytess: try: new_dict[key] = fmap(convert_bytes, value) except UnicodeDecodeError: if "decoding_failures" not in new_dict: new_dict["decoding_failures"] = [key] else: new_dict["decoding_failures"].append(key) new_dict[key] = fmap(decode_with_escape, value) elif key in empty_dict and not value: new_dict[key] = {} elif key in empty_list and not value: new_dict[key] = [] else: new_dict[key] = value _group_checksums(new_dict) return new_dict def from_origin(origin: Dict[str, Any]) -> OriginInfo: """Convert from a swh origin to an origin dictionary. """ return from_swh(origin) def from_release(release: Release) -> Dict[str, Any]: """Convert from a swh release to a json serializable release dictionary. Args: release: A release model object Returns: release dictionary with the following keys - id: hexadecimal sha1 (string) - revision: hexadecimal sha1 (string) - comment: release's comment message (string) - name: release's name (string) - author: release's author identifier (swh's id) - synthetic: the synthetic property (boolean) """ return from_swh( release.to_dict(), hashess={"id", "target"}, bytess={"message", "name", "fullname", "email"}, dates={"date"}, ) +class SWHDjangoJSONEncoder(DjangoJSONEncoder): + """Wrapper around DjangoJSONEncoder to serialize SWH-specific types + found in :class:`swh.web.common.typing.SWHObjectInfo`.""" + + def default(self, o): + if isinstance(o, ObjectType): + return o.name.lower() + else: + super().default(o) + + class SWHMetadataEncoder(json.JSONEncoder): """Special json encoder for metadata field which can contain bytes encoded value. """ def default(self, obj): if isinstance(obj, bytes): try: return obj.decode("utf-8") except UnicodeDecodeError: # fallback to binary representation to avoid display errors return repr(obj) # Let the base class default method raise the TypeError return json.JSONEncoder.default(self, obj) def convert_metadata(metadata): """Convert json specific dict to a json serializable one. """ if metadata is None: return {} return json.loads(json.dumps(metadata, cls=SWHMetadataEncoder)) def from_revision(revision: Union[Dict[str, Any], Revision]) -> Dict[str, Any]: """Convert swh revision model object to a json serializable revision dictionary. Args: revision: revision model object Returns: dict: Revision dictionary with the same keys as inputs, except: - sha1s are in hexadecimal strings (id, directory) - bytes are decoded in string (author_name, committer_name, author_email, committer_email) Remaining keys are left as is """ if isinstance(revision, Revision): revision_d = revision.to_dict() else: revision_d = revision revision_d = from_swh( revision_d, hashess={"id", "directory", "parents", "children"}, bytess={"name", "fullname", "email", "extra_headers", "message"}, convert={"metadata"}, convert_fn=convert_metadata, dates={"date", "committer_date"}, ) if revision_d: if "parents" in revision_d: revision_d["merge"] = len(revision_d["parents"]) > 1 return revision_d def from_raw_extrinsic_metadata( metadata: Union[Dict[str, Any], RawExtrinsicMetadata] ) -> Dict[str, Any]: """Convert RawExtrinsicMetadata model object to a json serializable dictionary. """ return from_swh( metadata.to_dict() if isinstance(metadata, RawExtrinsicMetadata) else metadata, blacklist={"id", "metadata"}, dates={"discovery_date"}, ) def from_content(content): """Convert swh content to serializable content dictionary. """ return from_swh( content, hashess={"sha1", "sha1_git", "sha256", "blake2s256"}, blacklist={"ctime"}, convert={"status"}, convert_fn=lambda v: "absent" if v == "hidden" else v, ) def from_person(person): """Convert swh person to serializable person dictionary. """ return from_swh(person, bytess={"name", "fullname", "email"}) def from_origin_visit(visit: Dict[str, Any]) -> OriginVisitInfo: """Convert swh origin_visit to serializable origin_visit dictionary. """ ov = from_swh( visit, hashess={"target", "snapshot"}, bytess={"branch"}, dates={"date"}, empty_dict={"metadata"}, ) return ov def from_snapshot(snapshot): """Convert swh snapshot to serializable (partial) snapshot dictionary. """ sv = from_swh(snapshot, hashess={"id", "target"}, bytess={"next_branch"}) if sv and "branches" in sv: sv["branches"] = {decode_with_escape(k): v for k, v in sv["branches"].items()} for k, v in snapshot["branches"].items(): # alias target existing branch names, not a sha1 if v and v["target_type"] == "alias": branch = decode_with_escape(k) target = decode_with_escape(v["target"]) sv["branches"][branch]["target"] = target return sv def from_partial_branches(branches: PartialBranches): """Convert PartialBranches to serializable partial snapshot dictionary """ return from_snapshot( { "id": branches["id"], "branches": { branch_name: branch.to_dict() if branch else None for (branch_name, branch) in branches["branches"].items() }, "next_branch": branches["next_branch"], } ) def from_directory_entry(dir_entry): """Convert swh directory to serializable directory dictionary. """ return from_swh( dir_entry, hashess={"dir_id", "sha1_git", "sha1", "sha256", "blake2s256", "target"}, bytess={"name"}, removables_if_empty={"sha1", "sha1_git", "sha256", "blake2s256", "status"}, convert={"status"}, convert_fn=lambda v: "absent" if v == "hidden" else v, ) def from_filetype(content_entry): """Convert swh content to serializable dictionary containing keys 'id', 'encoding', and 'mimetype'. """ return from_swh(content_entry, hashess={"id"}) diff --git a/swh/web/common/identifiers.py b/swh/web/common/identifiers.py index f295918e..43866f76 100644 --- a/swh/web/common/identifiers.py +++ b/swh/web/common/identifiers.py @@ -1,385 +1,388 @@ -# Copyright (C) 2020 The Software Heritage developers +# Copyright (C) 2020-2021 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 typing import Any, Dict, Iterable, List, Optional from urllib.parse import quote, unquote from typing_extensions import TypedDict from django.http import QueryDict from swh.model.exceptions import ValidationError from swh.model.hashutil import hash_to_bytes, hash_to_hex -from swh.model.identifiers import ( - CONTENT, - DIRECTORY, - RELEASE, - REVISION, - SNAPSHOT, - ObjectType, - QualifiedSWHID, -) +from swh.model.identifiers import ObjectType, QualifiedSWHID from swh.web.common import archive from swh.web.common.exc import BadInputExc from swh.web.common.typing import ( QueryParameters, SnapshotContext, SWHIDContext, SWHIDInfo, SWHObjectInfo, ) from swh.web.common.utils import reverse +def parse_object_type(object_type: str) -> ObjectType: + try: + return ObjectType[object_type.upper()] + except KeyError: + valid_types = ", ".join(variant.name.lower() for variant in ObjectType) + raise BadInputExc( + f"Invalid swh object type! Valid types are {valid_types}; not {object_type}" + ) + + def gen_swhid( - object_type: str, + object_type: ObjectType, object_id: str, scheme_version: int = 1, metadata: SWHIDContext = {}, ) -> str: """ Returns the SoftWare Heritage persistent IDentifier for a swh object based on: * the object type * the object id * the SWHID 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 SWHIDs Returns: the SWHID of the object Raises: BadInputExc: if the provided parameters do not enable to generate a valid identifier """ try: - decoded_object_type = ObjectType[object_type.upper()] decoded_object_id = hash_to_bytes(object_id) obj_swhid = str( QualifiedSWHID( - object_type=decoded_object_type, + object_type=object_type, object_id=decoded_object_id, scheme_version=scheme_version, **metadata, ) ) except (ValidationError, KeyError, ValueError) as e: raise BadInputExc("Invalid object (%s) for SWHID. %s" % (object_id, e)) else: return obj_swhid class ResolvedSWHID(TypedDict): """parsed SWHID with context""" swhid_parsed: QualifiedSWHID """URL to browse object according to SWHID context""" browse_url: Optional[str] def resolve_swhid( swhid: str, query_params: Optional[QueryParameters] = None ) -> ResolvedSWHID: """ Try to resolve a SoftWare Heritage persistent IDentifier into an url for browsing the targeted object. Args: swhid: 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: * **swhid_parsed**: the parsed identifier * **browse_url**: the url for browsing the targeted object """ swhid_parsed = get_swhid(swhid) object_type = swhid_parsed.object_type object_id = swhid_parsed.object_id browse_url = None url_args = {} query_dict = QueryDict("", mutable=True) fragment = "" process_lines = object_type == ObjectType.CONTENT if query_params and len(query_params) > 0: for k in sorted(query_params.keys()): query_dict[k] = query_params[k] if swhid_parsed.origin: origin_url = unquote(swhid_parsed.origin) origin_url = archive.lookup_origin({"url": origin_url})["url"] query_dict["origin_url"] = origin_url if swhid_parsed.path and swhid_parsed.path != b"/": query_dict["path"] = swhid_parsed.path.decode("utf8", errors="replace") if swhid_parsed.anchor: directory = b"" if swhid_parsed.anchor.object_type == ObjectType.DIRECTORY: directory = swhid_parsed.anchor.object_id elif swhid_parsed.anchor.object_type == ObjectType.REVISION: revision = archive.lookup_revision( hash_to_hex(swhid_parsed.anchor.object_id) ) directory = revision["directory"] elif swhid_parsed.anchor.object_type == ObjectType.RELEASE: release = archive.lookup_release( hash_to_hex(swhid_parsed.anchor.object_id) ) - if release["target_type"] == REVISION: + if release["target_type"] == ObjectType.REVISION.name.lower(): revision = archive.lookup_revision(release["target"]) directory = revision["directory"] if object_type == ObjectType.CONTENT: if ( not swhid_parsed.origin and swhid_parsed.anchor.object_type != ObjectType.REVISION ): # when no origin or revision context, content objects need to have # their path prefixed by root directory id for breadcrumbs display query_dict["path"] = hash_to_hex(directory) + query_dict["path"] else: # remove leading slash from SWHID content path query_dict["path"] = query_dict["path"][1:] elif object_type == ObjectType.DIRECTORY: object_id = directory # remove leading and trailing slashes from SWHID directory path if query_dict["path"].endswith("/"): query_dict["path"] = query_dict["path"][1:-1] else: query_dict["path"] = query_dict["path"][1:] # snapshot context if swhid_parsed.visit: if swhid_parsed.visit.object_type != ObjectType.SNAPSHOT: raise BadInputExc("Visit must be a snapshot SWHID.") query_dict["snapshot"] = hash_to_hex(swhid_parsed.visit.object_id) if swhid_parsed.anchor: if ( swhid_parsed.anchor.object_type == ObjectType.REVISION and object_type != ObjectType.REVISION ): query_dict["revision"] = hash_to_hex(swhid_parsed.anchor.object_id) elif swhid_parsed.anchor.object_type == ObjectType.RELEASE: release = archive.lookup_release( hash_to_hex(swhid_parsed.anchor.object_id) ) if release: query_dict["release"] = release["name"] # browsing content or directory without snapshot context elif ( object_type in (ObjectType.CONTENT, ObjectType.DIRECTORY) and swhid_parsed.anchor ): if swhid_parsed.anchor.object_type == ObjectType.REVISION: # anchor revision, objects are browsed from its view object_type = ObjectType.REVISION object_id = swhid_parsed.anchor.object_id elif ( object_type == ObjectType.DIRECTORY and swhid_parsed.anchor.object_type == ObjectType.DIRECTORY ): # a directory is browsed from its root object_id = swhid_parsed.anchor.object_id if object_type == ObjectType.CONTENT: url_args["query_string"] = f"sha1_git:{hash_to_hex(object_id)}" elif object_type in (ObjectType.DIRECTORY, ObjectType.RELEASE, ObjectType.REVISION): url_args["sha1_git"] = hash_to_hex(object_id) elif object_type == ObjectType.SNAPSHOT: url_args["snapshot_id"] = hash_to_hex(object_id) if swhid_parsed.lines and process_lines: lines = swhid_parsed.lines fragment += "#L" + str(lines[0]) if lines[1]: fragment += "-L" + str(lines[1]) if url_args: browse_url = ( reverse( f"browse-{object_type.name.lower()}", url_args=url_args, query_params=query_dict, ) + fragment ) return ResolvedSWHID(swhid_parsed=swhid_parsed, browse_url=browse_url) def get_swhid(swhid: str) -> QualifiedSWHID: """Check if a SWHID is valid and return it parsed. Args: swhid: a SoftWare Heritage persistent IDentifier. Raises: BadInputExc: if the provided SWHID can not be parsed. Return: A parsed SWHID. """ try: # ensure core part of SWHID is in lower case to avoid parsing error (core, sep, qualifiers) = swhid.partition(";") core = core.lower() return QualifiedSWHID.from_string(core + sep + qualifiers) except ValidationError as ve: raise BadInputExc("Error when parsing identifier: %s" % " ".join(ve.messages)) -def group_swhids(swhids: Iterable[QualifiedSWHID],) -> Dict[str, List[bytes]]: +def group_swhids(swhids: Iterable[QualifiedSWHID],) -> Dict[ObjectType, List[bytes]]: """ Groups many SoftWare Heritage persistent IDentifiers into a dictionary depending on their type. Args: swhids: an iterable of SoftWare Heritage persistent IDentifier objects Returns: A dictionary with: keys: object types values: object hashes """ - swhids_by_type: Dict[str, List[bytes]] = { - CONTENT: [], - DIRECTORY: [], - REVISION: [], - RELEASE: [], - SNAPSHOT: [], + swhids_by_type: Dict[ObjectType, List[bytes]] = { + ObjectType.CONTENT: [], + ObjectType.DIRECTORY: [], + ObjectType.REVISION: [], + ObjectType.RELEASE: [], + ObjectType.SNAPSHOT: [], } for obj_swhid in swhids: obj_id = obj_swhid.object_id obj_type = obj_swhid.object_type - swhids_by_type[obj_type.name.lower()].append(hash_to_bytes(obj_id)) + swhids_by_type[obj_type].append(hash_to_bytes(obj_id)) return swhids_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 SWHIDs of 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 SWHIDs 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: + if object_type != ObjectType.SNAPSHOT: swhid_context["visit"] = gen_swhid( - SNAPSHOT, snapshot_context["snapshot_id"] + ObjectType.SNAPSHOT, snapshot_context["snapshot_id"] ) - if object_type in (CONTENT, DIRECTORY): + if object_type in (ObjectType.CONTENT, ObjectType.DIRECTORY): if snapshot_context["release_id"] is not None: swhid_context["anchor"] = gen_swhid( - RELEASE, snapshot_context["release_id"] + ObjectType.RELEASE, snapshot_context["release_id"] ) elif snapshot_context["revision_id"] is not None: swhid_context["anchor"] = gen_swhid( - REVISION, snapshot_context["revision_id"] + ObjectType.REVISION, snapshot_context["revision_id"] ) - if object_type in (CONTENT, DIRECTORY): + if object_type in (ObjectType.CONTENT, ObjectType.DIRECTORY): if ( extra_context and "revision" in extra_context and extra_context["revision"] and "anchor" not in swhid_context ): - swhid_context["anchor"] = gen_swhid(REVISION, extra_context["revision"]) + swhid_context["anchor"] = gen_swhid( + ObjectType.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 + object_type != ObjectType.DIRECTORY or extra_context["root_directory"] != object_id ) ): swhid_context["anchor"] = gen_swhid( - DIRECTORY, extra_context["root_directory"] + ObjectType.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: + if "filename" in extra_context and object_type == ObjectType.CONTENT: path += extra_context["filename"] - if object_type == DIRECTORY and path == "/": + if object_type == ObjectType.DIRECTORY and path == "/": path = None if path: swhid_context["path"] = quote(path, safe="/?:@&") swhid = gen_swhid(object_type, object_id) swhid_url = reverse("browse-swhid", url_args={"swhid": swhid}) swhid_with_context = None swhid_with_context_url = None if swhid_context: swhid_with_context = gen_swhid( object_type, object_id, metadata=swhid_context ) swhid_with_context_url = reverse( "browse-swhid", url_args={"swhid": 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/swh_templatetags.py b/swh/web/common/swh_templatetags.py index 699b0c3b..1edbd632 100644 --- a/swh/web/common/swh_templatetags.py +++ b/swh/web/common/swh_templatetags.py @@ -1,153 +1,153 @@ # Copyright (C) 2017-2021 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 json import re from django import template -from django.core.serializers.json import DjangoJSONEncoder from django.utils.safestring import mark_safe +from swh.web.common.converters import SWHDjangoJSONEncoder from swh.web.common.origin_save import get_savable_visit_types from swh.web.common.utils import rst_to_html register = template.Library() @register.filter def docstring_display(docstring): """ Utility function to htmlize reST-formatted documentation in browsable api. """ return rst_to_html(docstring) @register.filter def urlize_links_and_mails(text): """Utility function for decorating api links in browsable api. Args: text: whose content matching links should be transformed into contextual API or Browse html links. Returns The text transformed if any link is found. The text as is otherwise. """ if 'href="' not in text: text = re.sub(r"(http.*)", r'\1', text) return re.sub(r'([^ <>"]+@[^ <>"]+)', r'\1', text) return text @register.filter def urlize_header_links(text): """Utility function for decorating headers links in browsable api. Args text: Text whose content contains Link header value Returns: The text transformed with html link if any link is found. The text as is otherwise. """ links = text.split(",") ret = "" for i, link in enumerate(links): ret += re.sub(r"<(http.*)>", r'<\1>', link) # add one link per line and align them if i != len(links) - 1: ret += "\n " return ret @register.filter def jsonify(obj): """Utility function for converting a django template variable to JSON in order to use it in script tags. Args obj: Any django template context variable Returns: JSON representation of the variable. """ - return mark_safe(json.dumps(obj, cls=DjangoJSONEncoder)) + return mark_safe(json.dumps(obj, cls=SWHDjangoJSONEncoder)) @register.filter def sub(value, arg): """Django template filter for subtracting two numbers Args: value (int/float): the value to subtract from arg (int/float): the value to subtract to Returns: int/float: The subtraction result """ return value - arg @register.filter def mul(value, arg): """Django template filter for multiplying two numbers Args: value (int/float): the value to multiply from arg (int/float): the value to multiply with Returns: int/float: The multiplication result """ return value * arg @register.filter def key_value(dict, key): """Django template filter to get a value in a dictionary. Args: dict (dict): a dictionary key (str): the key to lookup value Returns: The requested value in the dictionary """ return dict[key] @register.filter def visit_type_savable(visit_type: str) -> bool: """Django template filter to check if a save request can be created for a given visit type. Args: visit_type: the type of visit Returns: If the visit type is saveable or not """ return visit_type in get_savable_visit_types() @register.filter def split(value, arg): """Django template filter to split a string. Args: value (str): the string to split arg (str): the split separator Returns: list: the split string parts """ return value.split(arg) diff --git a/swh/web/common/typing.py b/swh/web/common/typing.py index 760e3daa..46866201 100644 --- a/swh/web/common/typing.py +++ b/swh/web/common/typing.py @@ -1,258 +1,259 @@ # Copyright (C) 2020-2021 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 typing import Any, Dict, List, Optional, TypeVar, Union from typing_extensions import TypedDict from django.http import QueryDict from swh.core.api.classes import PagedResult as CorePagedResult +from swh.model.identifiers import ObjectType QueryParameters = Union[Dict[str, Any], QueryDict] class OriginInfo(TypedDict): url: str """URL of the origin""" class OriginMetadataInfo(TypedDict): url: str """URL of the origin""" metadata: Dict[str, Any] """Origin metadata associated to the origin""" class OriginVisitInfo(TypedDict): date: str """date of the visit in iso format""" formatted_date: str """formatted date of the visit""" metadata: Dict[str, Any] """metadata associated to the visit""" origin: str """visited origin URL""" snapshot: str """snapshot identifier computed during the visit""" status: str """status of the visit ("ongoing", "full" or "partial") """ type: str """visit type (git, hg, debian, ...)""" url: str """URL to browse the snapshot""" visit: int """visit identifier""" class SnapshotBranchInfo(TypedDict): date: Optional[str] """"author date of branch heading revision""" directory: Optional[str] """directory associated to branch heading revision""" message: Optional[str] """message of branch heading revision""" name: str """branch name""" alias: bool """define if the branch is an alias""" revision: str """branch heading revision""" url: Optional[str] """optional browse URL (content, directory, ...) scoped to branch""" class SnapshotReleaseInfo(TypedDict): branch_name: str """branch name associated to release in snapshot""" date: str """release date""" directory: Optional[str] """optional directory associatd to the release""" id: str """release identifier""" message: str """release message""" name: str """release name""" alias: bool """define if the branch is an alias""" target: str """release target""" target_type: str """release target_type""" url: Optional[str] """optional browse URL (content, directory, ...) scoped to release""" class SnapshotContext(TypedDict): branch: Optional[str] """optional branch name set when browsing snapshot in that scope""" branch_alias: bool """indicates if the focused branch is an alias""" branches: List[SnapshotBranchInfo] """list of snapshot branches (possibly truncated)""" branches_url: str """snapshot branches list browse URL""" is_empty: bool """indicates if the snapshot is empty""" origin_info: Optional[OriginInfo] """optional origin info associated to the snapshot""" origin_visits_url: Optional[str] """optional origin visits URL""" query_params: QueryParameters """common query parameters when browsing snapshot content""" release: Optional[str] """optional release name set when browsing snapshot in that scope""" release_alias: bool """indicates if the focused release is an alias""" release_id: Optional[str] """optional release identifier set when browsing snapshot in that scope""" releases: List[SnapshotReleaseInfo] """list of snapshot releases (possibly truncated)""" releases_url: str """snapshot releases list browse URL""" revision_id: Optional[str] """optional revision identifier set when browsing snapshot in that scope""" revision_info: Optional[Dict[str, Any]] """optional revision info set when browsing snapshot in that scope""" root_directory: Optional[str] """optional root directory identifier set when browsing snapshot content""" snapshot_id: str """snapshot identifier""" snapshot_sizes: Dict[str, int] """snapshot sizes grouped by branch target type""" snapshot_swhid: str """snapshot SWHID""" url_args: Dict[str, Any] """common URL arguments when browsing snapshot content""" visit_info: Optional[OriginVisitInfo] """optional origin visit info associated to the snapshot""" class SWHObjectInfo(TypedDict): - object_type: str + object_type: ObjectType object_id: str class SWHIDContext(TypedDict, total=False): origin: str anchor: str visit: str path: str lines: str class SWHIDInfo(SWHObjectInfo): swhid: str swhid_url: str context: SWHIDContext swhid_with_context: Optional[str] swhid_with_context_url: Optional[str] class SWHObjectInfoMetadata(TypedDict, total=False): origin_url: Optional[str] visit_date: Optional[str] visit_type: Optional[str] class ContentMetadata(SWHObjectInfo, SWHObjectInfoMetadata): sha1: str sha1_git: str sha256: str blake2s256: str content_url: str mimetype: str encoding: str size: int language: str path: Optional[str] filename: Optional[str] directory: Optional[str] root_directory: Optional[str] revision: Optional[str] release: Optional[str] snapshot: Optional[str] class DirectoryMetadata(SWHObjectInfo, SWHObjectInfoMetadata): directory: str nb_files: int nb_dirs: int sum_file_sizes: int root_directory: Optional[str] path: str revision: Optional[str] revision_found: Optional[bool] release: Optional[str] snapshot: Optional[str] class ReleaseMetadata(SWHObjectInfo, SWHObjectInfoMetadata): release: str author: str author_url: str date: str name: str synthetic: bool target: str target_type: str snapshot: Optional[str] class RevisionMetadata(SWHObjectInfo, SWHObjectInfoMetadata): revision: str author: str author_url: str committer: str committer_url: str date: str committer_date: str directory: str merge: bool metadata: str parents: List[str] synthetic: bool type: str snapshot: Optional[str] TResult = TypeVar("TResult") PagedResult = CorePagedResult[TResult, str] class SaveOriginRequestInfo(TypedDict): id: int """Unique key""" save_request_date: str """Date of the creation request""" visit_type: str """Type of the visit""" visit_status: Optional[str] """Status of the visit""" origin_url: str """Origin to ingest""" save_request_status: str """Status of the request""" loading_task_id: Optional[int] """Identifier of the loading task in the scheduler if scheduled""" visit_date: Optional[str] """End of the visit if terminated""" save_task_status: str """Status of the scheduled task""" class OriginExistenceCheckInfo(TypedDict): origin_url: str """Origin to check""" exists: bool """Does the url exist?""" content_length: Optional[int] """content length of the artifact""" last_modified: Optional[str] """Last modification time reported by the server (as iso8601 string)""" diff --git a/swh/web/misc/badges.py b/swh/web/misc/badges.py index 16d6da66..a6e30d3e 100644 --- a/swh/web/misc/badges.py +++ b/swh/web/misc/badges.py @@ -1,166 +1,168 @@ -# Copyright (C) 2019-2020 The Software Heritage developers +# Copyright (C) 2019-2021 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 base64 import b64encode from typing import Optional, cast from pybadges import badge from django.conf.urls import url from django.contrib.staticfiles import finders from django.http import HttpRequest, HttpResponse from swh.model.exceptions import ValidationError from swh.model.hashutil import hash_to_bytes, hash_to_hex -from swh.model.identifiers import RELEASE, CoreSWHID, ObjectType, QualifiedSWHID +from swh.model.identifiers import CoreSWHID, ObjectType, QualifiedSWHID from swh.web.common import archive from swh.web.common.exc import BadInputExc, NotFoundExc -from swh.web.common.identifiers import resolve_swhid +from swh.web.common.identifiers import parse_object_type, resolve_swhid from swh.web.common.utils import reverse _orange = "#f36a24" _blue = "#0172b2" _red = "#cd5741" _swh_logo_data = None _badge_config = { "content": {"color": _blue, "title": "Archived source file",}, "directory": {"color": _blue, "title": "Archived source tree",}, "origin": {"color": _orange, "title": "Archived software repository",}, "release": {"color": _blue, "title": "Archived software release",}, "revision": {"color": _blue, "title": "Archived commit",}, "snapshot": {"color": _blue, "title": "Archived software repository snapshot",}, "error": {"color": _red, "title": "An error occurred when generating the badge"}, } def _get_logo_data() -> str: """ Get data-URI for Software Heritage SVG logo to embed it in the generated badges. """ global _swh_logo_data if _swh_logo_data is None: swh_logo_path = cast(str, finders.find("img/swh-logo-white.svg")) with open(swh_logo_path, "rb") as swh_logo_file: _swh_logo_data = "data:image/svg+xml;base64,%s" % b64encode( swh_logo_file.read() ).decode("ascii") return _swh_logo_data def _swh_badge( request: HttpRequest, object_type: str, object_id: str, object_swhid: Optional[str] = "", ) -> HttpResponse: """ Generate a Software Heritage badge for a given object type and id. Args: request: input http request object_type: The type of swh object to generate a badge for, either *content*, *directory*, *revision*, *release*, *origin* or *snapshot* object_id: The id of the swh object, either an url for origin type or a *sha1* for other object types object_swhid: If provided, the object SWHID will not be recomputed Returns: HTTP response with content type *image/svg+xml* containing the SVG badge data. If the provided parameters are invalid, HTTP 400 status code will be returned. If the object can not be found in the archive, HTTP 404 status code will be returned. """ left_text = "error" whole_link = None try: if object_type == "origin": archive.lookup_origin({"url": object_id}) right_text = "repository" whole_link = reverse( "browse-origin", query_params={"origin_url": object_id} ) else: # when SWHID is provided, object type and id will be parsed # from it if object_swhid: parsed_swhid = QualifiedSWHID.from_string(object_swhid) - object_type = parsed_swhid.object_type.name.lower() + parsed_object_type = parsed_swhid.object_type object_id = hash_to_hex(parsed_swhid.object_id) - swh_object = archive.lookup_object(object_type, object_id) + swh_object = archive.lookup_object(parsed_swhid.object_type, object_id) # remove SWHID qualified if any for badge text right_text = str( CoreSWHID( object_type=parsed_swhid.object_type, object_id=parsed_swhid.object_id, ) ) + object_type = parsed_swhid.object_type.name.lower() else: + parsed_object_type = parse_object_type(object_type) right_text = str( CoreSWHID( - object_type=ObjectType[object_type.upper()], + object_type=parsed_object_type, object_id=hash_to_bytes(object_id), ) ) - swh_object = archive.lookup_object(object_type, object_id) + swh_object = archive.lookup_object(parsed_object_type, object_id) whole_link = resolve_swhid(str(right_text))["browse_url"] # use release name for badge text - if object_type == RELEASE: + if parsed_object_type == ObjectType.RELEASE: right_text = "release %s" % swh_object["name"] left_text = "archived" except (BadInputExc, ValidationError): right_text = f'invalid {object_type if object_type else "object"} id' object_type = "error" except NotFoundExc: right_text = f'{object_type if object_type else "object"} not found' object_type = "error" badge_data = badge( left_text=left_text, right_text=right_text, right_color=_badge_config[object_type]["color"], whole_link=request.build_absolute_uri(whole_link), whole_title=_badge_config[object_type]["title"], logo=_get_logo_data(), embed_logo=True, ) return HttpResponse(badge_data, content_type="image/svg+xml") def _swh_badge_swhid(request: HttpRequest, object_swhid: str) -> HttpResponse: """ Generate a Software Heritage badge for a given object SWHID. Args: request (django.http.HttpRequest): input http request object_swhid (str): a SWHID of an archived object Returns: django.http.HttpResponse: An http response with content type *image/svg+xml* containing the SVG badge data. If any error occurs, a status code of 400 will be returned. """ return _swh_badge(request, "", "", object_swhid) urlpatterns = [ url( r"^badge/(?P[a-z]+)/(?P.+)/$", _swh_badge, name="swh-badge", ), url( r"^badge/(?Pswh:[0-9]+:[a-z]+:[0-9a-f]+.*)/$", _swh_badge_swhid, name="swh-badge-swhid", ), ] diff --git a/swh/web/misc/iframe.py b/swh/web/misc/iframe.py index e52e033a..5c19f98c 100644 --- a/swh/web/misc/iframe.py +++ b/swh/web/misc/iframe.py @@ -1,310 +1,306 @@ # Copyright (C) 2021 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 typing import Any, Dict, List, Optional from django.conf.urls import url from django.shortcuts import render from django.views.decorators.clickjacking import xframe_options_exempt from swh.model.hashutil import hash_to_bytes -from swh.model.identifiers import ( - CONTENT, - DIRECTORY, - REVISION, - SNAPSHOT, - ObjectType, - QualifiedSWHID, -) +from swh.model.identifiers import ObjectType, QualifiedSWHID from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import ( content_display_max_size, get_directory_entries, prepare_content_for_display, request_content, ) from swh.web.common import archive from swh.web.common.exc import BadInputExc, NotFoundExc, http_status_code_message from swh.web.common.identifiers import get_swhid, get_swhids_info from swh.web.common.typing import SnapshotContext, SWHObjectInfo from swh.web.common.utils import gen_path_info, reverse def _get_content_rendering_data(cnt_swhid: QualifiedSWHID, path: str) -> Dict[str, Any]: content_data = request_content(f"sha1_git:{cnt_swhid.object_id.hex()}") content = None language = None mimetype = None if content_data.get("raw_data") is not None: content_display_data = prepare_content_for_display( content_data["raw_data"], content_data["mimetype"], path ) content = content_display_data["content_data"] language = content_display_data["language"] mimetype = content_display_data["mimetype"] return { "content": content, "content_size": content_data.get("length"), "max_content_size": content_display_max_size, "filename": path.split("/")[-1], "encoding": content_data.get("encoding"), "mimetype": mimetype, "language": language, } def _get_directory_rendering_data( dir_swhid: QualifiedSWHID, focus_swhid: QualifiedSWHID, path: str, ) -> Dict[str, Any]: dirs, files = get_directory_entries(dir_swhid.object_id.hex()) for d in dirs: if d["type"] == "rev": d["url"] = None else: dir_swhid = QualifiedSWHID( object_type=ObjectType.DIRECTORY, object_id=hash_to_bytes(d["target"]), origin=dir_swhid.origin, visit=dir_swhid.visit, anchor=dir_swhid.anchor, path=(path or "/") + d["name"] + "/", ) d["url"] = reverse( "swhid-iframe", url_args={"swhid": str(dir_swhid)}, query_params={"focus_swhid": str(focus_swhid)}, ) for f in files: object_id = hash_to_bytes(f["target"]) cnt_swhid = QualifiedSWHID( object_type=ObjectType.CONTENT, object_id=object_id, origin=dir_swhid.origin, visit=dir_swhid.visit, anchor=dir_swhid.anchor, path=(path or "/") + f["name"], lines=(focus_swhid.lines if object_id == focus_swhid.object_id else None), ) f["url"] = reverse( "swhid-iframe", url_args={"swhid": str(cnt_swhid)}, query_params={"focus_swhid": str(focus_swhid)}, ) return {"dirs": dirs, "files": files} def _get_breacrumbs_data( swhid: QualifiedSWHID, focus_swhid: QualifiedSWHID, path: str, snapshot_context: Optional[SnapshotContext] = None, ) -> List[Dict[str, Any]]: breadcrumbs = [] filename = None # strip any leading or trailing slash from path qualifier of SWHID if path and path[0] == "/": path = path[1:] if path and path[-1] == "/": path = path[:-1] if swhid.object_type == ObjectType.CONTENT: split_path = path.split("/") filename = split_path[-1] path = path[: -len(filename)] path_info = gen_path_info(path) if path != "/" else [] root_dir = None if snapshot_context and snapshot_context["root_directory"]: root_dir = snapshot_context["root_directory"] elif focus_swhid.object_type == ObjectType.DIRECTORY: root_dir = focus_swhid.object_id.hex() if root_dir: root_dir_swhid = QualifiedSWHID( object_type=ObjectType.DIRECTORY, object_id=hash_to_bytes(root_dir), origin=swhid.origin, visit=swhid.visit, anchor=swhid.anchor, ) breadcrumbs.append( { "name": root_dir[:7], "object_id": root_dir_swhid.object_id.hex(), "path": "/", "url": reverse( "swhid-iframe", url_args={"swhid": str(root_dir_swhid)}, query_params={"focus_swhid": focus_swhid}, ), } ) for pi in path_info: dir_info = archive.lookup_directory_with_path(root_dir, pi["path"]) dir_swhid = QualifiedSWHID( object_type=ObjectType.DIRECTORY, object_id=hash_to_bytes(dir_info["target"]), origin=swhid.origin, visit=swhid.visit, anchor=swhid.anchor, path="/" + pi["path"] + "/", ) breadcrumbs.append( { "name": pi["name"], "object_id": dir_swhid.object_id.hex(), "path": dir_swhid.path.decode("utf-8") if dir_swhid.path else "", "url": reverse( "swhid-iframe", url_args={"swhid": str(dir_swhid)}, query_params={"focus_swhid": focus_swhid}, ), } ) if filename: breadcrumbs.append( { "name": filename, "object_id": swhid.object_id.hex(), "path": path, "url": "", } ) return breadcrumbs @xframe_options_exempt def swhid_iframe(request, swhid: str): """Django view that can be embedded in an iframe to display objects archived by Software Heritage (currently contents and directories) in a minimalist Web UI. """ focus_swhid = request.GET.get("focus_swhid", swhid) parsed_swhid = None view_data = {} breadcrumbs = [] swh_objects = [] snapshot_context = None swhids_info_extra_context = {} try: parsed_swhid = get_swhid(swhid) parsed_focus_swhid = get_swhid(focus_swhid) path = parsed_swhid.path.decode("utf-8") if parsed_swhid.path else "" snapshot_context = None revision_id = None if ( parsed_swhid.anchor and parsed_swhid.anchor.object_type == ObjectType.REVISION ): revision_id = parsed_swhid.anchor.object_id.hex() if parsed_swhid.origin or parsed_swhid.visit: snapshot_context = get_snapshot_context( origin_url=parsed_swhid.origin, snapshot_id=parsed_swhid.visit.object_id.hex() if parsed_swhid.visit else None, revision_id=revision_id, ) error_info: Dict[str, Any] = {"status_code": 200, "description": ""} if parsed_swhid and parsed_swhid.object_type == ObjectType.CONTENT: view_data = _get_content_rendering_data(parsed_swhid, path) swh_objects.append( SWHObjectInfo( - object_type=CONTENT, object_id=parsed_swhid.object_id.hex() + object_type=ObjectType.CONTENT, + object_id=parsed_swhid.object_id.hex(), ) ) elif parsed_swhid and parsed_swhid.object_type == ObjectType.DIRECTORY: view_data = _get_directory_rendering_data( parsed_swhid, parsed_focus_swhid, path ) swh_objects.append( SWHObjectInfo( - object_type=DIRECTORY, object_id=parsed_swhid.object_id.hex() + object_type=ObjectType.DIRECTORY, + object_id=parsed_swhid.object_id.hex(), ) ) elif parsed_swhid: error_info = { "status_code": 400, "description": ( f"Objects of type {parsed_swhid.object_type} are not supported" ), } swhids_info_extra_context["path"] = path if parsed_swhid and view_data: breadcrumbs = _get_breacrumbs_data( parsed_swhid, parsed_focus_swhid, path, snapshot_context ) if parsed_swhid.object_type == ObjectType.CONTENT and len(breadcrumbs) > 1: swh_objects.append( SWHObjectInfo( - object_type=DIRECTORY, object_id=breadcrumbs[-2]["object_id"] + object_type=ObjectType.DIRECTORY, + object_id=breadcrumbs[-2]["object_id"], ) ) swhids_info_extra_context["path"] = breadcrumbs[-2]["path"] swhids_info_extra_context["filename"] = breadcrumbs[-1]["name"] if snapshot_context: swh_objects.append( SWHObjectInfo( - object_type=REVISION, + object_type=ObjectType.REVISION, object_id=snapshot_context["revision_id"] or "", ) ) swh_objects.append( SWHObjectInfo( - object_type=SNAPSHOT, + object_type=ObjectType.SNAPSHOT, object_id=snapshot_context["snapshot_id"] or "", ) ) except BadInputExc as e: error_info = {"status_code": 400, "description": f"BadInputExc: {str(e)}"} except NotFoundExc as e: error_info = {"status_code": 404, "description": f"NotFoundExc: {str(e)}"} except Exception as e: error_info = {"status_code": 500, "description": str(e)} return render( request, "misc/iframe.html", { **view_data, "iframe_mode": True, "object_type": parsed_swhid.object_type.value if parsed_swhid else None, "lines": parsed_swhid.lines if parsed_swhid else None, "breadcrumbs": breadcrumbs, "swhid": swhid, "focus_swhid": focus_swhid, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], "snapshot_context": None, "swhids_info": get_swhids_info( swh_objects, snapshot_context, swhids_info_extra_context ), }, status=error_info["status_code"], ) urlpatterns = [ url( r"^embed/(?Pswh:[0-9]+:[a-z]+:[0-9a-f]+.*)$", swhid_iframe, name="swhid-iframe", ), ] diff --git a/swh/web/templates/includes/show-metadata.html b/swh/web/templates/includes/show-metadata.html index 5f2a146a..18c25095 100644 --- a/swh/web/templates/includes/show-metadata.html +++ b/swh/web/templates/includes/show-metadata.html @@ -1,52 +1,52 @@ {% comment %} -Copyright (C) 2017-2020 The Software Heritage developers +Copyright (C) 2017-2021 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 {% endcomment %} {% load swh_templatetags %} {% if 'metadata' in swh_object_metadata and swh_object_metadata.metadata != '{}' %} {% endif %} {% if user.is_staff %} {% if swhids_info %}
{% for swhid_info in swhids_info %} - + {{swhid_info.object_type}} {% endfor %}
{% endif %} {% endif %} diff --git a/swh/web/templates/includes/show-swhids.html b/swh/web/templates/includes/show-swhids.html index fe533eb2..7012a4b3 100644 --- a/swh/web/templates/includes/show-swhids.html +++ b/swh/web/templates/includes/show-swhids.html @@ -1,118 +1,118 @@ {% comment %} Copyright (C) 2017-2021 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 {% endcomment %} {% load swh_templatetags %} {% if swhids_info %}