diff --git a/swh/web/browse/browseurls.py b/swh/web/browse/browseurls.py index 08afd147..6cfd9e6f 100644 --- a/swh/web/browse/browseurls.py +++ b/swh/web/browse/browseurls.py @@ -1,42 +1,48 @@ -# Copyright (C) 2017-2019 The Software Heritage developers +# Copyright (C) 2017-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information +from typing import List, Optional + from swh.web.common.urlsindex import UrlsIndex class BrowseUrls(UrlsIndex): """ Class to manage swh-web browse application urls. """ scope = "browse" -def browse_route(*url_patterns, view_name=None, checksum_args=None): +def browse_route( + *url_patterns: str, + view_name: Optional[str] = None, + checksum_args: Optional[List[str]] = None, +): """ Decorator to ease the registration of a swh-web browse endpoint Args: url_patterns: list of url patterns used by Django to identify the browse routes view_name: the name of the Django view associated to the routes used to reverse the url """ - url_patterns = ["^" + url_pattern + "$" for url_pattern in url_patterns] + url_patterns = tuple("^" + url_pattern + "$" for url_pattern in url_patterns) view_name = view_name def decorator(f): # register the route and its view in the browse endpoints index for url_pattern in url_patterns: BrowseUrls.add_url_pattern(url_pattern, f, view_name) if checksum_args: BrowseUrls.add_redirect_for_checksum_args( view_name, url_patterns, checksum_args ) return f return decorator diff --git a/swh/web/browse/identifiers.py b/swh/web/browse/identifiers.py index 885b47b8..f096a250 100644 --- a/swh/web/browse/identifiers.py +++ b/swh/web/browse/identifiers.py @@ -1,19 +1,20 @@ -# Copyright (C) 2017-2020 The Software Heritage developers +# Copyright (C) 2017-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information +from django.http import HttpRequest, HttpResponse from django.shortcuts import redirect from swh.web.common.identifiers import resolve_swhid -def swhid_browse(request, swhid): +def swhid_browse(request: HttpRequest, swhid: str) -> HttpResponse: """ Django view enabling to browse the archive using :ref:`persistent-identifiers`. The url that points to it is :http:get:`/(swhid)/`. """ swhid_resolved = resolve_swhid(swhid, query_params=request.GET) - + assert swhid_resolved["browse_url"] return redirect(swhid_resolved["browse_url"]) diff --git a/swh/web/browse/snapshot_context.py b/swh/web/browse/snapshot_context.py index a44f7580..c4e8a5d0 100644 --- a/swh/web/browse/snapshot_context.py +++ b/swh/web/browse/snapshot_context.py @@ -1,1335 +1,1379 @@ # Copyright (C) 2018-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information # Utility module for browsing the archive in a snapshot context. from collections import defaultdict from typing import Any, Dict, List, Optional, Tuple +from django.http import HttpRequest, HttpResponse from django.shortcuts import render from django.utils.html import escape from swh.model.hashutil import hash_to_bytes from swh.model.model import Snapshot from swh.model.swhids import CoreSWHID, ObjectType from swh.web.browse.utils import ( format_log_entries, gen_release_link, gen_revision_link, gen_revision_log_link, gen_revision_url, gen_snapshot_link, get_directory_entries, get_readme_to_display, ) from swh.web.common import archive from swh.web.common.exc import BadInputExc, NotFoundExc, http_status_code_message from swh.web.common.identifiers import get_swhids_info from swh.web.common.origin_visits import get_origin_visit from swh.web.common.typing import ( DirectoryMetadata, OriginInfo, SnapshotBranchInfo, SnapshotContext, SnapshotReleaseInfo, SWHObjectInfo, ) from swh.web.common.utils import ( django_cache, 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): +def _get_branch( + branches: List[SnapshotBranchInfo], branch_name: str, snapshot_id: str +) -> Optional[SnapshotBranchInfo]: """ Utility function to get a specific branch from a snapshot. Returns None if the branch cannot be found. """ filtered_branches = [b for b in branches if b["name"] == branch_name] if filtered_branches: return filtered_branches[0] else: # case where a large branches list has been truncated snp = archive.lookup_snapshot( snapshot_id, branches_from=branch_name, branches_count=1, target_types=["revision", "alias"], # pull request branches must be browsable even if they are hidden # by default in branches list branch_name_exclude_prefix=None, ) snp_branch, _, _ = process_snapshot_branches(snp) if snp_branch and snp_branch[0]["name"] == branch_name: branches.append(snp_branch[0]) return snp_branch[0] + return None -def _get_release(releases, release_name, snapshot_id): +def _get_release( + releases: List[SnapshotReleaseInfo], release_name: Optional[str], snapshot_id: str +) -> Optional[SnapshotReleaseInfo]: """ Utility function to get a specific release from a snapshot. Returns None if the release cannot be found. """ filtered_releases = [r for r in releases if r["name"] == release_name] if filtered_releases: return filtered_releases[0] - else: + elif release_name: # case where a large branches list has been truncated try: # git origins have specific branches for releases snp = archive.lookup_snapshot( snapshot_id, branches_from=f"refs/tags/{release_name}", branches_count=1, target_types=["release"], ) except NotFoundExc: snp = archive.lookup_snapshot( snapshot_id, branches_from=release_name, branches_count=1, target_types=["release", "alias"], ) _, snp_release, _ = process_snapshot_branches(snp) if snp_release and snp_release[0]["name"] == release_name: releases.append(snp_release[0]) return snp_release[0] + return None def _branch_not_found( - branch_type, branch, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id -): + branch_type: str, + branch: str, + snapshot_id: str, + snapshot_sizes: Dict[str, int], + origin_info: Optional[OriginInfo], + timestamp: Optional[str], + visit_id: Optional[int], +) -> None: """ Utility function to raise an exception when a specified branch/release can not be found. """ if branch_type == "branch": branch_type = "Branch" branch_type_plural = "branches" 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: + elif visit_id and snapshot_sizes[target_type] == 0 and origin_info: 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: + elif visit_id and origin_info: 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: + elif snapshot_sizes[target_type] == 0 and origin_info and timestamp: msg = ( "Origin with url %s" " for visit with timestamp %s has an empty list" " of %s!" % (origin_info["url"], timestamp, branch_type_plural) ) - else: + elif origin_info and timestamp: 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 @django_cache() 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 """ 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) return branches, releases, aliases def get_origin_visit_snapshot( origin_info: OriginInfo, visit_ts: Optional[str] = None, visit_id: Optional[int] = None, snapshot_id: Optional[str] = None, ) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo], Dict[str, Any]]: """Returns the lists of branches and releases associated to an origin for a given visit. The visit is expressed by either: * a snapshot identifier * a timestamp, if no visit with that exact timestamp is found, the closest one from the provided timestamp will be used. If no visit parameter is provided, it returns the list of branches found for the latest visit. That list is put in cache in order to speedup the navigation in the swh-web/browse ui. .. warning:: At most 1000 branches contained in the snapshot will be returned for performance reasons. Args: origin_info: a dict filled with origin information visit_ts: an ISO 8601 datetime string to parse visit_id: visit id for disambiguation in case several visits have the same timestamp snapshot_id: if provided, visit associated to the snapshot will be processed Returns: A tuple with three members. The first one is a list of dict describing the origin branches for the given visit. The second one is a list of dict describing the origin releases for the given visit. The third one is a dict mapping resolved branch aliases to their real target. Raises: NotFoundExc if the origin or its visit are not found """ visit_info = get_origin_visit(origin_info, visit_ts, visit_id, snapshot_id) return get_snapshot_content(visit_info["snapshot"]) def get_snapshot_context( snapshot_id: Optional[str] = None, origin_url: Optional[str] = None, timestamp: Optional[str] = None, visit_id: Optional[int] = None, branch_name: Optional[str] = None, release_name: Optional[str] = None, revision_id: Optional[str] = None, path: Optional[str] = None, browse_context: str = "directory", ) -> SnapshotContext: """ Utility function to compute relevant information when navigating the archive in a snapshot context. The snapshot is either referenced by its id or it will be retrieved from an origin visit. Args: snapshot_id: hexadecimal representation of a snapshot identifier origin_url: an origin_url timestamp: a datetime string for retrieving the closest visit of the origin visit_id: optional visit id for disambiguation in case of several visits with the same timestamp branch_name: optional branch name set when browsing the snapshot in that scope (will default to "HEAD" if not provided) release_name: optional release name set when browsing the snapshot in that scope revision_id: optional revision identifier set when browsing the snapshot in that scope path: optional path of the object currently browsed in the snapshot browse_context: indicates which type of object is currently browsed Returns: A dict filled with snapshot context information. Raises: swh.web.common.exc.NotFoundExc: if no snapshot is found for the visit of an origin. """ assert origin_url is not None or snapshot_id is not None origin_info = None visit_info = None url_args = {} query_params: Dict[str, Any] = {} origin_visits_url = None if origin_url: if visit_id is not None: query_params["visit_id"] = visit_id elif snapshot_id is not None: query_params["snapshot"] = snapshot_id origin_info = archive.lookup_origin({"url": origin_url}) visit_info = get_origin_visit(origin_info, timestamp, visit_id, snapshot_id) formatted_date = format_utc_iso_date(visit_info["date"]) visit_info["formatted_date"] = formatted_date snapshot_id = visit_info["snapshot"] if not snapshot_id: raise NotFoundExc( "No snapshot associated to the visit of origin " "%s on %s" % (escape(origin_url), formatted_date) ) # provided timestamp is not necessarily equals to the one # of the retrieved visit, so get the exact one in order # to use it in the urls generated below if timestamp: timestamp = visit_info["date"] branches, releases, aliases = get_origin_visit_snapshot( origin_info, timestamp, visit_id, snapshot_id ) query_params["origin_url"] = origin_info["url"] origin_visits_url = reverse( "browse-origin-visits", query_params={"origin_url": origin_info["url"]} ) if timestamp is not None: query_params["timestamp"] = format_utc_iso_date( timestamp, "%Y-%m-%dT%H:%M:%SZ" ) visit_url = reverse("browse-origin-directory", query_params=query_params) visit_info["url"] = directory_url = visit_url branches_url = reverse("browse-origin-branches", query_params=query_params) releases_url = reverse("browse-origin-releases", query_params=query_params) else: assert snapshot_id is not None branches, releases, aliases = get_snapshot_content(snapshot_id) url_args = {"snapshot_id": snapshot_id} directory_url = reverse("browse-snapshot-directory", url_args=url_args) branches_url = reverse("browse-snapshot-branches", url_args=url_args) releases_url = reverse("browse-snapshot-releases", url_args=url_args) releases = list(reversed(releases)) @django_cache() def _get_snapshot_sizes(snapshot_id): return archive.lookup_snapshot_sizes(snapshot_id) snapshot_sizes = _get_snapshot_sizes(snapshot_id) is_empty = (snapshot_sizes["release"] + snapshot_sizes["revision"]) == 0 swh_snp_id = str( CoreSWHID(object_type=ObjectType.SNAPSHOT, object_id=hash_to_bytes(snapshot_id)) ) if visit_info: timestamp = format_utc_iso_date(visit_info["date"]) if origin_info: browse_view_name = f"browse-origin-{browse_context}" else: browse_view_name = f"browse-snapshot-{browse_context}" release_id = None root_directory = None snapshot_total_size = snapshot_sizes["release"] + snapshot_sizes["revision"] if path is not None: query_params["path"] = path if snapshot_total_size and revision_id is not None: # browse specific revision for a snapshot requested revision = archive.lookup_revision(revision_id) root_directory = revision["directory"] branches.append( SnapshotBranchInfo( name=revision_id, alias=False, revision=revision_id, directory=root_directory, date=revision["date"], message=revision["message"], url=None, ) ) query_params["revision"] = revision_id elif snapshot_total_size and release_name: # browse specific release for a snapshot requested release = _get_release(releases, release_name, snapshot_id) if release is None: _branch_not_found( "release", release_name, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id, ) else: if release["target_type"] == "revision": revision = archive.lookup_revision(release["target"]) root_directory = revision["directory"] revision_id = release["target"] elif release["target_type"] == "directory": root_directory = release["target"] release_id = release["id"] query_params["release"] = release_name elif snapshot_total_size: head = aliases.get("HEAD") if branch_name: # browse specific branch for a snapshot requested query_params["branch"] = branch_name branch = _get_branch(branches, branch_name, snapshot_id) if branch is None: _branch_not_found( "branch", branch_name, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id, ) else: branch_name = branch["name"] revision_id = branch["revision"] root_directory = branch["directory"] elif head is not None: # otherwise, browse branch targeted by the HEAD alias if it exists if head["target_type"] == "revision": # HEAD alias targets a revision head_rev = archive.lookup_revision(head["target"]) branch_name = "HEAD" revision_id = head_rev["id"] root_directory = head_rev["directory"] else: # HEAD alias targets a release release_name = archive.lookup_release(head["target"])["name"] head_rel = _get_release(releases, release_name, snapshot_id) - if head_rel["target_type"] == "revision": + if head_rel is None: + _branch_not_found( + "release", + str(release_name), + snapshot_id, + snapshot_sizes, + origin_info, + timestamp, + visit_id, + ) + elif head_rel["target_type"] == "revision": revision = archive.lookup_revision(head_rel["target"]) root_directory = revision["directory"] revision_id = head_rel["target"] elif head_rel["target_type"] == "directory": root_directory = head_rel["target"] - release_id = head_rel["id"] + if head_rel is not None: + release_id = head_rel["id"] elif branches: # fallback to browse first branch otherwise branch = branches[0] branch_name = branch["name"] revision_id = branch["revision"] root_directory = branch["directory"] elif releases: # fallback to browse last release otherwise release = releases[-1] if release["target_type"] == "revision": revision = archive.lookup_revision(release["target"]) root_directory = revision["directory"] revision_id = release["target"] elif release["target_type"] == "directory": root_directory = release["target"] release_id = release["id"] release_name = release["name"] for b in branches: branch_query_params = dict(query_params) branch_query_params.pop("release", None) if b["name"] != b["revision"]: branch_query_params.pop("revision", None) branch_query_params["branch"] = b["name"] b["url"] = reverse( browse_view_name, url_args=url_args, query_params=branch_query_params ) for r in releases: release_query_params = dict(query_params) release_query_params.pop("branch", None) release_query_params.pop("revision", None) release_query_params["release"] = r["name"] r["url"] = reverse( browse_view_name, url_args=url_args, query_params=release_query_params, ) revision_info = None if revision_id: try: revision_info = archive.lookup_revision(revision_id) except NotFoundExc: pass else: revision_info["date"] = format_utc_iso_date(revision_info["date"]) revision_info["committer_date"] = format_utc_iso_date( revision_info["committer_date"] ) if revision_info["message"]: message_lines = revision_info["message"].split("\n") revision_info["message_header"] = message_lines[0] else: revision_info["message_header"] = "" snapshot_context = SnapshotContext( directory_url=directory_url, branch=branch_name, branch_alias=branch_name in aliases, branches=branches, branches_url=branches_url, is_empty=is_empty, origin_info=origin_info, origin_visits_url=origin_visits_url, release=release_name, release_alias=release_name in aliases, release_id=release_id, query_params=query_params, releases=releases, releases_url=releases_url, revision_id=revision_id, revision_info=revision_info, root_directory=root_directory, snapshot_id=snapshot_id, snapshot_sizes=snapshot_sizes, snapshot_swhid=swh_snp_id, url_args=url_args, visit_info=visit_info, ) if revision_info: - revision_info["revision_url"] = gen_revision_url(revision_id, snapshot_context) + revision_info["revision_url"] = gen_revision_url( + revision_info["id"], snapshot_context + ) return snapshot_context -def _build_breadcrumbs(snapshot_context: SnapshotContext, path: str): +def _build_breadcrumbs( + snapshot_context: SnapshotContext, path: Optional[str] +) -> List[Dict[str, 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): +def _check_origin_url(snapshot_id: Optional[str], origin_url: Optional[str]) -> None: 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 -): + request: HttpRequest, + snapshot_id: Optional[str] = None, + origin_url: Optional[str] = None, + timestamp: Optional[str] = None, + path: Optional[str] = None, +) -> HttpResponse: """ Django view implementation for browsing a directory in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) visit_id = int(request.GET.get("visit_id", 0)) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=visit_id or None, path=path, browse_context="directory", branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), ) root_directory = snapshot_context["root_directory"] sha1_git = root_directory - error_info = { + error_info: Dict[str, Any] = { "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 = { + vault_cooking: Dict[str, Any] = { "directory_context": False, "directory_swhid": None, "revision_context": False, "revision_swhid": None, } revision_found = False if revision_id is not None: try: archive.lookup_revision(revision_id) except NotFoundExc: pass else: revision_found = True if sha1_git is not None: swh_objects.append( SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=sha1_git) ) vault_cooking.update( { "directory_context": True, "directory_swhid": f"swh:1:dir:{sha1_git}", } ) if revision_id is not None and revision_found: swh_objects.append( SWHObjectInfo(object_type=ObjectType.REVISION, object_id=revision_id) ) vault_cooking.update( { "revision_context": True, "revision_swhid": f"swh:1:rev:{revision_id}", } ) swh_objects.append( SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id) ) visit_date = None visit_type = None if visit_info: visit_date = format_utc_iso_date(visit_info["date"]) visit_type = visit_info["type"] release_id = snapshot_context["release_id"] if release_id: swh_objects.append( SWHObjectInfo(object_type=ObjectType.RELEASE, object_id=release_id) ) dir_metadata = DirectoryMetadata( object_type=ObjectType.DIRECTORY, object_id=sha1_git, directory=sha1_git, nb_files=nb_files, nb_dirs=nb_dirs, sum_file_sizes=sum_file_sizes, root_directory=root_directory, path=dir_path, revision=revision_id, revision_found=revision_found, release=release_id, snapshot=snapshot_id, origin_url=origin_url, visit_date=visit_date, visit_type=visit_type, ) swhids_info = get_swhids_info(swh_objects, snapshot_context, dir_metadata) dir_path = "/".join([bc["name"] for bc in breadcrumbs]) + "/" context_found = "snapshot: %s" % snapshot_context["snapshot_id"] if origin_info: context_found = "origin: %s" % origin_info["url"] heading = "Directory - %s - %s - %s" % ( dir_path, snapshot_context["branch"], context_found, ) top_right_link = None if not snapshot_context["is_empty"] and revision_found: top_right_link = { "url": history_url, "icon": swh_object_icons["revisions history"], "text": "History", } return render( request, "browse/directory.html", { "heading": heading, "swh_object_name": "Directory", "swh_object_metadata": dir_metadata, "dirs": dirs, "files": files, "breadcrumbs": breadcrumbs if root_directory else [], "top_right_link": top_right_link, "readme_name": readme_name, "readme_url": readme_url, "readme_html": readme_html, "snapshot_context": snapshot_context, "vault_cooking": vault_cooking, "show_actions": True, "swhids_info": swhids_info, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], }, status=error_info["status_code"], ) PER_PAGE = 100 -def browse_snapshot_log(request, snapshot_id=None, origin_url=None, timestamp=None): +def browse_snapshot_log( + request: HttpRequest, + snapshot_id: Optional[str] = None, + origin_url: Optional[str] = None, + timestamp: Optional[str] = None, +) -> HttpResponse: """ Django view implementation for browsing a revision history in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) visit_id = int(request.GET.get("visit_id", 0)) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=visit_id or None, browse_context="log", branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), ) revision_id = snapshot_context["revision_id"] if revision_id is None: raise NotFoundExc("No revisions history found in the current snapshot context.") per_page = int(request.GET.get("per_page", PER_PAGE)) offset = int(request.GET.get("offset", 0)) revs_ordering = request.GET.get("revs_ordering", "committer_date") session_key = "rev_%s_log_ordering_%s" % (revision_id, revs_ordering) rev_log_session = request.session.get(session_key, None) rev_log = [] revs_walker_state = None if rev_log_session: rev_log = rev_log_session["rev_log"] revs_walker_state = rev_log_session["revs_walker_state"] if len(rev_log) < offset + per_page: revs_walker = archive.get_revisions_walker( revs_ordering, revision_id, max_revs=offset + per_page + 1, state=revs_walker_state, ) rev_log += [rev["id"] for rev in revs_walker] revs_walker_state = revs_walker.export_state() revs = rev_log[offset : offset + per_page] revision_log = archive.lookup_revision_multiple(revs) request.session[session_key] = { "rev_log": rev_log, "revs_walker_state": revs_walker_state, } origin_info = snapshot_context["origin_info"] visit_info = snapshot_context["visit_info"] url_args = snapshot_context["url_args"] query_params = snapshot_context["query_params"] snapshot_id = snapshot_context["snapshot_id"] - query_params["per_page"] = per_page + query_params["per_page"] = str(per_page) revs_ordering = request.GET.get("revs_ordering", "") - query_params["revs_ordering"] = revs_ordering or None + if revs_ordering: + query_params["revs_ordering"] = revs_ordering 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 + query_params["offset"] = str(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 + query_params["offset"] = str(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: + if origin_info and visit_info: revision_metadata["origin url"] = origin_info["url"] revision_metadata["origin visit date"] = format_utc_iso_date(visit_info["date"]) revision_metadata["origin visit type"] = visit_info["type"] swh_objects = [ SWHObjectInfo(object_type=ObjectType.REVISION, object_id=revision_id), SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id), ] release_id = snapshot_context["release_id"] if release_id: swh_objects.append( SWHObjectInfo(object_type=ObjectType.RELEASE, object_id=release_id) ) browse_rel_link = gen_release_link(release_id) revision_metadata["release"] = release_id revision_metadata["context-independent release"] = browse_rel_link swhids_info = get_swhids_info(swh_objects, snapshot_context) context_found = "snapshot: %s" % snapshot_context["snapshot_id"] if origin_info: context_found = "origin: %s" % origin_info["url"] heading = "Revision history - %s - %s" % (snapshot_context["branch"], context_found) return render( request, "browse/revision-log.html", { "heading": heading, "swh_object_name": "Revisions history", "swh_object_metadata": revision_metadata, "revision_log": revision_log_data, "revs_ordering": revs_ordering, "next_log_url": next_log_url, "prev_log_url": prev_log_url, "breadcrumbs": None, "top_right_link": None, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": True, "swhids_info": swhids_info, }, ) def browse_snapshot_branches( - request, snapshot_id=None, origin_url=None, timestamp=None, branch_name_include=None -): + request: HttpRequest, + snapshot_id: Optional[str] = None, + origin_url: Optional[str] = None, + timestamp: Optional[str] = None, + branch_name_include: Optional[str] = None, +) -> HttpResponse: """ Django view implementation for browsing a list of branches in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) visit_id = int(request.GET.get("visit_id", 0)) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=visit_id or None, ) - branches_bc = request.GET.get("branches_breadcrumbs", "") - branches_bc = branches_bc.split(",") if branches_bc else [] + branches_bc_str = request.GET.get("branches_breadcrumbs", "") + branches_bc = branches_bc_str.split(",") if branches_bc_str 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 = [] + displayed_branches: List[Dict[str, Any]] = [] if snapshot: - displayed_branches, _, _ = process_snapshot_branches(snapshot) + branches, _, _ = process_snapshot_branches(snapshot) + displayed_branches = [dict(branch) for branch in branches] 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, + request: HttpRequest, + snapshot_id: Optional[str] = None, + origin_url: Optional[str] = None, + timestamp: Optional[str] = None, + release_name_include: Optional[str] = None, ): """ Django view implementation for browsing a list of releases in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) visit_id = int(request.GET.get("visit_id", 0)) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=visit_id or None, ) - rel_bc = request.GET.get("releases_breadcrumbs", "") - rel_bc = rel_bc.split(",") if rel_bc else [] + rel_bc_str = request.GET.get("releases_breadcrumbs", "") + rel_bc = rel_bc_str.split(",") if rel_bc_str 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 = [] + displayed_releases: List[Dict[str, Any]] = [] if snapshot: - _, displayed_releases, _ = process_snapshot_branches(snapshot) + _, releases, _ = process_snapshot_branches(snapshot) + displayed_releases = [dict(release) for release in releases] for release in displayed_releases: query_params_tgt = {"snapshot": snapshot_id, "release": release["name"]} if origin_info: query_params_tgt["origin_url"] = origin_info["url"] release_url = reverse( "browse-release", url_args={"sha1_git": release["id"]}, query_params=query_params_tgt, ) target_url = "" tooltip = ( f"The release {release['name']} targets " f"{release['target_type']} {release['target']}" ) if release["target_type"] == "revision": target_url = reverse( "browse-revision", url_args={"sha1_git": release["target"]}, query_params=query_params_tgt, ) elif release["target_type"] == "directory": target_url = reverse( "browse-directory", url_args={"sha1_git": release["target"]}, query_params=query_params_tgt, ) elif release["target_type"] == "content": target_url = reverse( "browse-content", url_args={"query_string": release["target"]}, query_params=query_params_tgt, ) elif release["target_type"] == "release": target_url = reverse( "browse-release", url_args={"sha1_git": release["target"]}, query_params=query_params_tgt, ) tooltip = ( f"The release {release['name']} " f"is an alias for release {release['target']}" ) release["release_url"] = release_url release["target_url"] = target_url release["tooltip"] = tooltip if origin_info: browse_view_name = "browse-origin-releases" else: browse_view_name = "browse-snapshot-releases" prev_releases_url = None next_releases_url = None if rel_bc: query_params_prev = dict(query_params) query_params_prev["releases_breadcrumbs"] = ",".join(rel_bc[:-1]) prev_releases_url = reverse( browse_view_name, url_args=url_args, query_params=query_params_prev ) elif rel_from: prev_releases_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) if snapshot and snapshot["next_branch"] is not None: query_params_next = dict(query_params) next_rel = displayed_releases[-1]["branch_name"] del displayed_releases[-1] rel_bc.append(next_rel) query_params_next["releases_breadcrumbs"] = ",".join(rel_bc) next_releases_url = reverse( browse_view_name, url_args=url_args, query_params=query_params_next ) heading = "Releases - " if origin_info: heading += "origin: %s" % origin_info["url"] else: heading += "snapshot: %s" % snapshot_id return render( request, "browse/releases.html", { "heading": heading, "top_panel_visible": False, "top_panel_collapsible": False, "swh_object_name": "Releases", "swh_object_metadata": {}, "top_right_link": None, "displayed_releases": displayed_releases, "prev_releases_url": prev_releases_url, "next_releases_url": next_releases_url, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": False, "search_string": release_name_include or "", }, ) diff --git a/swh/web/browse/urls.py b/swh/web/browse/urls.py index 4dad5d3b..92efd002 100644 --- a/swh/web/browse/urls.py +++ b/swh/web/browse/urls.py @@ -1,63 +1,64 @@ -# Copyright (C) 2017-2021 The Software Heritage developers +# Copyright (C) 2017-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from django.conf.urls import url +from django.http import HttpRequest, HttpResponse from django.shortcuts import redirect, render from swh.web.browse.browseurls import BrowseUrls from swh.web.browse.identifiers import swhid_browse import swh.web.browse.views.content # noqa import swh.web.browse.views.directory # noqa import swh.web.browse.views.origin # noqa import swh.web.browse.views.release # noqa import swh.web.browse.views.revision # noqa import swh.web.browse.views.snapshot # noqa from swh.web.common.utils import origin_visit_types, reverse -def _browse_help_view(request): +def _browse_help_view(request: HttpRequest) -> HttpResponse: return render( request, "browse/help.html", {"heading": "How to browse the archive ?"} ) -def _browse_search_view(request): +def _browse_search_view(request: HttpRequest) -> HttpResponse: return render( request, "browse/search.html", { "heading": "Search software origins to browse", "visit_types": origin_visit_types(), }, ) -def _browse_vault_view(request): +def _browse_vault_view(request: HttpRequest) -> HttpResponse: return render( request, "browse/vault-ui.html", {"heading": "Download archive content from the Vault"}, ) -def _browse_origin_save_view(request): +def _browse_origin_save_view(request: HttpRequest) -> HttpResponse: return redirect(reverse("origin-save")) urlpatterns = [ url(r"^$", _browse_search_view), url(r"^help/$", _browse_help_view, name="browse-help"), url(r"^search/$", _browse_search_view, name="browse-search"), url(r"^vault/$", _browse_vault_view, name="browse-vault"), # for backward compatibility url(r"^origin/save/$", _browse_origin_save_view, name="browse-origin-save"), url( r"^(?Pswh:[0-9]+:[a-z]+:[0-9a-f]+.*)/$", swhid_browse, name="browse-swhid", ), ] urlpatterns += BrowseUrls.get_url_patterns() diff --git a/swh/web/browse/utils.py b/swh/web/browse/utils.py index cfac337b..a39bfbf4 100644 --- a/swh/web/browse/utils.py +++ b/swh/web/browse/utils.py @@ -1,725 +1,773 @@ # Copyright (C) 2017-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import base64 import stat import textwrap -from typing import Tuple +from typing import Any, Dict, Iterator, List, Optional, Tuple, Union import chardet import magic from django.utils.html import escape from django.utils.safestring import mark_safe from swh.web.common import archive, highlightjs from swh.web.common.exc import NotFoundExc, sentry_capture_exception +from swh.web.common.typing import SnapshotContext from swh.web.common.utils import ( browsers_supported_image_mimes, django_cache, format_utc_iso_date, reverse, rst_to_html, ) from swh.web.config import get_config @django_cache() -def get_directory_entries(sha1_git): +def get_directory_entries( + sha1_git: str, +) -> Tuple[List[Dict[str, Any]], List[Dict[str, Any]]]: """Function that retrieves the content of a directory from the archive. The directories entries are first sorted in lexicographical order. Sub-directories and regular files are then extracted. Args: sha1_git: sha1_git identifier of the directory Returns: A tuple whose first member corresponds to the sub-directories list and second member the regular files list Raises: NotFoundExc if the directory is not found """ - entries = list(archive.lookup_directory(sha1_git)) + entries: List[Dict[str, Any]] = list(archive.lookup_directory(sha1_git)) for e in entries: e["perms"] = stat.filemode(e["perms"]) if e["type"] == "rev": # modify dir entry name to explicitly show it points # to a revision e["name"] = "%s @ %s" % (e["name"], e["target"][:7]) dirs = [e for e in entries if e["type"] in ("dir", "rev")] files = [e for e in entries if e["type"] == "file"] dirs = sorted(dirs, key=lambda d: d["name"]) files = sorted(files, key=lambda f: f["name"]) return dirs, files -def get_mimetype_and_encoding_for_content(content): +def get_mimetype_and_encoding_for_content(content: bytes) -> Tuple[str, str]: """Function that returns the mime type and the encoding associated to a content buffer using the magic module under the hood. Args: - content (bytes): a content buffer + content: a content buffer Returns: A tuple (mimetype, encoding), for instance ('text/plain', 'us-ascii'), associated to the provided content. """ m = magic.Magic(mime=True, mime_encoding=True) mime_encoding = m.from_buffer(content) mime_type, encoding = mime_encoding.split(";") encoding = encoding.replace(" charset=", "") return mime_type, encoding # maximum authorized content size in bytes for HTML display # with code highlighting content_display_max_size = get_config()["content_display_max_size"] def re_encode_content( mimetype: str, encoding: str, content_data: bytes ) -> Tuple[str, str, bytes]: """Try to re-encode textual content if it is not encoded to UTF-8 for proper display in the browse Web UI. Args: mimetype: content mimetype as detected by python-magic encoding: content encoding as detected by python-magic content_data: raw content bytes Returns: A tuple with 3 members: content mimetype, content encoding (possibly updated after processing), content raw bytes (possibly reencoded to UTF-8) """ if mimetype.startswith("text/") and encoding not in ("us-ascii", "utf-8"): # first check if chardet detects an encoding with confidence result = chardet.detect(content_data) if result["confidence"] >= 0.9: encoding = result["encoding"] content_data = content_data.decode(encoding).encode("utf-8") elif encoding == "unknown-8bit": # probably a malformed UTF-8 content, re-encode it # by replacing invalid chars with a substitution one content_data = content_data.decode("utf-8", "replace").encode("utf-8") elif encoding not in ["utf-8", "binary"]: content_data = content_data.decode(encoding, "replace").encode("utf-8") elif mimetype.startswith("application/octet-stream"): # file may detect a text content as binary # so try to decode it for display encodings = ["us-ascii", "utf-8"] encodings += ["iso-8859-%s" % i for i in range(1, 17)] for enc in encodings: try: content_data = content_data.decode(enc).encode("utf-8") except Exception: pass else: # ensure display in content view encoding = enc mimetype = "text/plain" break return mimetype, encoding, content_data def request_content( - query_string, - max_size=content_display_max_size, - re_encode=True, -): + query_string: str, + max_size: Optional[int] = content_display_max_size, + re_encode: bool = True, +) -> Dict[str, Any]: """Function that retrieves a content from the archive. Raw bytes content is first retrieved, then the content mime type. If the mime type is not stored in the archive, it will be computed using Python magic module. Args: query_string: a string of the form "[ALGO_HASH:]HASH" where optional ALGO_HASH can be either ``sha1``, ``sha1_git``, ``sha256``, or ``blake2s256`` (default to ``sha1``) and HASH the hexadecimal representation of the hash value max_size: the maximum size for a content to retrieve (default to 1MB, no size limit if None) Returns: - A tuple whose first member corresponds to the content raw bytes - and second member the content mime type + A dict filled with content info. Raises: NotFoundExc if the content is not found """ content_data = archive.lookup_content(query_string) filetype = None language = None # requests to the indexer db may fail so properly handle # those cases in order to avoid content display errors try: filetype = archive.lookup_content_filetype(query_string) language = archive.lookup_content_language(query_string) except Exception as exc: sentry_capture_exception(exc) mimetype = "unknown" encoding = "unknown" if filetype: mimetype = filetype["mimetype"] encoding = filetype["encoding"] if not max_size or content_data["length"] < max_size: try: content_raw = archive.lookup_content_raw(query_string) except Exception as exc: sentry_capture_exception(exc) raise NotFoundExc( "The bytes of the content are currently not available " "in the archive." ) else: content_data["raw_data"] = content_raw["data"] if not filetype: mimetype, encoding = get_mimetype_and_encoding_for_content( content_data["raw_data"] ) if re_encode: mimetype, encoding, raw_data = re_encode_content( mimetype, encoding, content_data["raw_data"] ) content_data["raw_data"] = raw_data else: content_data["raw_data"] = None content_data["mimetype"] = mimetype content_data["encoding"] = encoding if language: content_data["language"] = language["lang"] else: content_data["language"] = "not detected" return content_data -def prepare_content_for_display(content_data, mime_type, path): +def prepare_content_for_display( + content_data: bytes, mime_type: str, path: Optional[str] +) -> Dict[str, Any]: """Function that prepares a content for HTML display. The function tries to associate a programming language to a content in order to perform syntax highlighting client-side using highlightjs. The language is determined using either the content filename or its mime type. If the mime type corresponds to an image format supported by web browsers, the content will be encoded in base64 for displaying the image. Args: - content_data (bytes): raw bytes of the content - mime_type (string): mime type of the content - path (string): path of the content including filename + content_data: raw bytes of the content + mime_type: mime type of the content + path: path of the content including filename Returns: A dict containing the content bytes (possibly different from the one provided as parameter if it is an image) under the key 'content_data and the corresponding highlightjs language class under the key 'language'. """ language = None if path: language = highlightjs.get_hljs_language_from_filename(path.split("/")[-1]) if language is None: language = highlightjs.get_hljs_language_from_mime_type(mime_type) if language is None: language = "plaintext" + processed_content: Union[bytes, str] = content_data + if mime_type.startswith("image/"): if mime_type in browsers_supported_image_mimes: - content_data = base64.b64encode(content_data).decode("ascii") + processed_content = base64.b64encode(content_data).decode("ascii") if mime_type.startswith("image/svg"): mime_type = "image/svg+xml" if mime_type.startswith("text/") or mime_type.startswith("application/"): - content_data = content_data.decode("utf-8", errors="replace") + processed_content = content_data.decode("utf-8", errors="replace") - return {"content_data": content_data, "language": language, "mimetype": mime_type} + return { + "content_data": processed_content, + "language": language, + "mimetype": mime_type, + } -def gen_link(url, link_text=None, link_attrs=None): +def gen_link( + url: str, + link_text: Optional[str] = None, + link_attrs: Optional[Dict[str, str]] = None, +) -> str: """ Utility function for generating an HTML link to insert in Django templates. Args: - url (str): an url - link_text (str): optional text for the produced link, + url: an url + link_text: optional text for the produced link, if not provided the url will be used - link_attrs (dict): optional attributes (e.g. class) + link_attrs: optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ attrs = " " if link_attrs: for k, v in link_attrs.items(): attrs += '%s="%s" ' % (k, v) if not link_text: link_text = url link = '%s' % (attrs, escape(url), escape(link_text)) return mark_safe(link) -def _snapshot_context_query_params(snapshot_context): - query_params = {} +def _snapshot_context_query_params( + snapshot_context: Optional[SnapshotContext], +) -> Dict[str, str]: + query_params: Dict[str, str] = {} if not snapshot_context: return query_params if snapshot_context and snapshot_context["origin_info"]: origin_info = snapshot_context["origin_info"] snp_query_params = snapshot_context["query_params"] query_params = {"origin_url": origin_info["url"]} if "timestamp" in snp_query_params: - query_params["timestamp"] = snp_query_params["timestamp"] + query_params["timestamp"] = str(snp_query_params["timestamp"]) if "visit_id" in snp_query_params: - query_params["visit_id"] = snp_query_params["visit_id"] + query_params["visit_id"] = str(snp_query_params["visit_id"]) if "snapshot" in snp_query_params and "visit_id" not in query_params: - query_params["snapshot"] = snp_query_params["snapshot"] + query_params["snapshot"] = str(snp_query_params["snapshot"]) elif snapshot_context: query_params = {"snapshot": snapshot_context["snapshot_id"]} if snapshot_context["release"]: query_params["release"] = snapshot_context["release"] elif snapshot_context["branch"] and snapshot_context["branch"] not in ( "HEAD", snapshot_context["revision_id"], ): query_params["branch"] = snapshot_context["branch"] elif snapshot_context["revision_id"]: query_params["revision"] = snapshot_context["revision_id"] return query_params -def gen_revision_url(revision_id, snapshot_context=None): +def gen_revision_url( + revision_id: str, snapshot_context: Optional[SnapshotContext] = None +) -> str: """ Utility function for generating an url to a revision. Args: - revision_id (str): a revision id - snapshot_context (dict): if provided, generate snapshot-dependent + revision_id: a revision id + snapshot_context: if provided, generate snapshot-dependent browsing url Returns: str: The url to browse the revision """ query_params = _snapshot_context_query_params(snapshot_context) # remove query parameters not needed for a revision view query_params.pop("revision", None) query_params.pop("release", None) return reverse( "browse-revision", url_args={"sha1_git": revision_id}, query_params=query_params ) def gen_revision_link( - revision_id, - shorten_id=False, - snapshot_context=None, - link_text="Browse", - link_attrs={"class": "btn btn-default btn-sm", "role": "button"}, -): + revision_id: str, + shorten_id: bool = False, + snapshot_context: Optional[SnapshotContext] = None, + link_text: Optional[str] = "Browse", + link_attrs: Optional[Dict[str, str]] = { + "class": "btn btn-default btn-sm", + "role": "button", + }, +) -> Optional[str]: """ Utility function for generating a link to a revision HTML view to insert in Django templates. Args: - revision_id (str): a revision id - shorten_id (boolean): whether to shorten the revision id to 7 + revision_id: a revision id + shorten_id: whether to shorten the revision id to 7 characters for the link text - snapshot_context (dict): if provided, generate snapshot-dependent + snapshot_context: if provided, generate snapshot-dependent browsing link - link_text (str): optional text for the generated link + link_text: optional text for the generated link (the revision id will be used by default) - link_attrs (dict): optional attributes (e.g. class) + link_attrs: optional attributes (e.g. class) to add to the link Returns: str: An HTML link in the form 'revision_id' """ if not revision_id: return None revision_url = gen_revision_url(revision_id, snapshot_context) if shorten_id: return gen_link(revision_url, revision_id[:7], link_attrs) else: if not link_text: link_text = revision_id return gen_link(revision_url, link_text, link_attrs) def gen_directory_link( - sha1_git, - snapshot_context=None, - link_text="Browse", - link_attrs={"class": "btn btn-default btn-sm", "role": "button"}, -): + sha1_git: str, + snapshot_context: Optional[SnapshotContext] = None, + link_text: Optional[str] = "Browse", + link_attrs: Optional[Dict[str, str]] = { + "class": "btn btn-default btn-sm", + "role": "button", + }, +) -> Optional[str]: """ Utility function for generating a link to a directory HTML view to insert in Django templates. Args: - sha1_git (str): directory identifier - link_text (str): optional text for the generated link + sha1_git: directory identifier + link_text: optional text for the generated link (the directory id will be used by default) - link_attrs (dict): optional attributes (e.g. class) + link_attrs: optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ if not sha1_git: return None query_params = _snapshot_context_query_params(snapshot_context) directory_url = reverse( "browse-directory", url_args={"sha1_git": sha1_git}, query_params=query_params ) if not link_text: link_text = sha1_git return gen_link(directory_url, link_text, link_attrs) def gen_snapshot_link( - snapshot_id, - snapshot_context=None, - link_text="Browse", - link_attrs={"class": "btn btn-default btn-sm", "role": "button"}, -): + snapshot_id: str, + snapshot_context: Optional[SnapshotContext] = None, + link_text: Optional[str] = "Browse", + link_attrs: Optional[Dict[str, str]] = { + "class": "btn btn-default btn-sm", + "role": "button", + }, +) -> str: """ Utility function for generating a link to a snapshot HTML view to insert in Django templates. Args: - snapshot_id (str): snapshot identifier - link_text (str): optional text for the generated link + snapshot_id: snapshot identifier + link_text: optional text for the generated link (the snapshot id will be used by default) - link_attrs (dict): optional attributes (e.g. class) + link_attrs: optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ query_params = _snapshot_context_query_params(snapshot_context) snapshot_url = reverse( "browse-snapshot", url_args={"snapshot_id": snapshot_id}, query_params=query_params, ) if not link_text: link_text = snapshot_id return gen_link(snapshot_url, link_text, link_attrs) def gen_content_link( - sha1_git, - snapshot_context=None, - link_text="Browse", - link_attrs={"class": "btn btn-default btn-sm", "role": "button"}, -): + sha1_git: str, + snapshot_context: Optional[SnapshotContext] = None, + link_text: Optional[str] = "Browse", + link_attrs: Optional[Dict[str, str]] = { + "class": "btn btn-default btn-sm", + "role": "button", + }, +) -> Optional[str]: """ Utility function for generating a link to a content HTML view to insert in Django templates. Args: - sha1_git (str): content identifier - link_text (str): optional text for the generated link + sha1_git: content identifier + link_text: optional text for the generated link (the content sha1_git will be used by default) - link_attrs (dict): optional attributes (e.g. class) + link_attrs: optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ if not sha1_git: return None query_params = _snapshot_context_query_params(snapshot_context) content_url = reverse( "browse-content", url_args={"query_string": "sha1_git:" + sha1_git}, query_params=query_params, ) if not link_text: link_text = sha1_git return gen_link(content_url, link_text, link_attrs) -def get_revision_log_url(revision_id, snapshot_context=None): +def get_revision_log_url( + revision_id: str, snapshot_context: Optional[SnapshotContext] = None +) -> str: """ Utility function for getting the URL for a revision log HTML view (possibly in the context of an origin). Args: - revision_id (str): revision identifier the history heads to - snapshot_context (dict): if provided, generate snapshot-dependent + revision_id: revision identifier the history heads to + snapshot_context: if provided, generate snapshot-dependent browsing link Returns: The revision log view URL """ query_params = {} if snapshot_context: query_params = _snapshot_context_query_params(snapshot_context) query_params["revision"] = revision_id if snapshot_context and snapshot_context["origin_info"]: revision_log_url = reverse("browse-origin-log", query_params=query_params) elif snapshot_context: url_args = {"snapshot_id": snapshot_context["snapshot_id"]} del query_params["snapshot"] revision_log_url = reverse( "browse-snapshot-log", url_args=url_args, query_params=query_params ) else: revision_log_url = reverse( "browse-revision-log", url_args={"sha1_git": revision_id} ) return revision_log_url def gen_revision_log_link( - revision_id, - snapshot_context=None, - link_text="Browse", - link_attrs={"class": "btn btn-default btn-sm", "role": "button"}, -): + revision_id: str, + snapshot_context: Optional[SnapshotContext] = None, + link_text: Optional[str] = "Browse", + link_attrs: Optional[Dict[str, str]] = { + "class": "btn btn-default btn-sm", + "role": "button", + }, +) -> Optional[str]: """ Utility function for generating a link to a revision log HTML view (possibly in the context of an origin) to insert in Django templates. Args: - revision_id (str): revision identifier the history heads to - snapshot_context (dict): if provided, generate snapshot-dependent + revision_id: revision identifier the history heads to + snapshot_context: if provided, generate snapshot-dependent browsing link - link_text (str): optional text to use for the generated link + link_text: optional text to use for the generated link (the revision id will be used by default) - link_attrs (dict): optional attributes (e.g. class) + link_attrs: optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ if not revision_id: return None revision_log_url = get_revision_log_url(revision_id, snapshot_context) if not link_text: link_text = revision_id return gen_link(revision_log_url, link_text, link_attrs) -def gen_person_mail_link(person, link_text=None): +def gen_person_mail_link( + person: Dict[str, Any], link_text: Optional[str] = None +) -> str: """ Utility function for generating a mail link to a person to insert in Django templates. Args: - person (dict): dictionary containing person data + person: dictionary containing person data (*name*, *email*, *fullname*) - link_text (str): optional text to use for the generated mail link + link_text: optional text to use for the generated mail link (the person name will be used by default) Returns: str: A mail link to the person or the person name if no email is present in person data """ person_name = person["name"] or person["fullname"] or "None" if link_text is None: link_text = person_name person_email = person["email"] if person["email"] else None if person_email is None and "@" in person_name and " " not in person_name: person_email = person_name if person_email: return gen_link(url="mailto:%s" % person_email, link_text=link_text) else: return person_name def gen_release_link( - sha1_git, - snapshot_context=None, - link_text="Browse", - link_attrs={"class": "btn btn-default btn-sm", "role": "button"}, -): + sha1_git: str, + snapshot_context: Optional[SnapshotContext] = None, + link_text: Optional[str] = "Browse", + link_attrs: Optional[Dict[str, str]] = { + "class": "btn btn-default btn-sm", + "role": "button", + }, +) -> str: """ Utility function for generating a link to a release HTML view to insert in Django templates. Args: - sha1_git (str): release identifier - link_text (str): optional text for the generated link + sha1_git: release identifier + link_text: optional text for the generated link (the release id will be used by default) - link_attrs (dict): optional attributes (e.g. class) + link_attrs: optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ query_params = _snapshot_context_query_params(snapshot_context) release_url = reverse( "browse-release", url_args={"sha1_git": sha1_git}, query_params=query_params ) if not link_text: link_text = sha1_git return gen_link(release_url, link_text, link_attrs) -def format_log_entries(revision_log, per_page, snapshot_context=None): +def format_log_entries( + revision_log: Iterator[Optional[Dict[str, Any]]], + per_page: int, + snapshot_context: Optional[SnapshotContext] = None, +) -> List[Dict[str, str]]: """ Utility functions that process raw revision log data for HTML display. Its purpose is to: * add links to relevant browse views * format date in human readable format * truncate the message log Args: - revision_log (list): raw revision log as returned by the swh-web api - per_page (int): number of log entries per page - snapshot_context (dict): if provided, generate snapshot-dependent + revision_log: raw revision log as returned by the swh-web api + per_page: number of log entries per page + snapshot_context: if provided, generate snapshot-dependent browsing link """ revision_log_data = [] for i, rev in enumerate(revision_log): + if rev is None: + continue if i == per_page: break author_name = "None" author_fullname = "None" committer_fullname = "None" if rev["author"]: author_name = gen_person_mail_link(rev["author"]) author_fullname = rev["author"]["fullname"] if rev["committer"]: committer_fullname = rev["committer"]["fullname"] author_date = format_utc_iso_date(rev["date"]) committer_date = format_utc_iso_date(rev["committer_date"]) tooltip = "revision %s\n" % rev["id"] tooltip += "author: %s\n" % author_fullname tooltip += "author date: %s\n" % author_date tooltip += "committer: %s\n" % committer_fullname tooltip += "committer date: %s\n\n" % committer_date if rev["message"]: tooltip += textwrap.indent(rev["message"], " " * 4) revision_log_data.append( { "author": author_name, "id": rev["id"][:7], "message": rev["message"], "date": author_date, "commit_date": committer_date, "url": gen_revision_url(rev["id"], snapshot_context), "tooltip": tooltip, } ) return revision_log_data # list of common readme names ordered by preference # (lower indices have higher priority) _common_readme_names = [ "readme.markdown", "readme.md", "readme.rst", "readme.txt", "readme", ] -def get_readme_to_display(readmes): +def get_readme_to_display( + readmes: Dict[str, str] +) -> Tuple[Optional[str], Optional[str], Optional[str]]: """ Process a list of readme files found in a directory in order to find the adequate one to display. Args: - readmes: a list of dict where keys are readme file names and values + readmes: a dict where keys are readme file names and values are readme sha1s Returns: - A tuple (readme_name, readme_sha1) + A tuple (readme_name, readme_url, readme_html) """ readme_name = None readme_url = None readme_sha1 = None readme_html = None lc_readmes = {k.lower(): {"orig_name": k, "sha1": v} for k, v in readmes.items()} # look for readme names according to the preference order # defined by the _common_readme_names list for common_readme_name in _common_readme_names: if common_readme_name in lc_readmes: readme_name = lc_readmes[common_readme_name]["orig_name"] readme_sha1 = lc_readmes[common_readme_name]["sha1"] readme_url = reverse( "browse-content-raw", url_args={"query_string": readme_sha1}, query_params={"re_encode": "true"}, ) break # otherwise pick the first readme like file if any if not readme_name and len(readmes.items()) > 0: readme_name = next(iter(readmes)) readme_sha1 = readmes[readme_name] readme_url = reverse( "browse-content-raw", url_args={"query_string": readme_sha1}, query_params={"re_encode": "true"}, ) # convert rst README to html server side as there is # no viable solution to perform that task client side if readme_name and readme_name.endswith(".rst"): @django_cache( catch_exception=True, exception_return_value="Readme bytes are not available", ) def _rst_readme_to_html(readme_sha1): rst_doc = request_content(readme_sha1) return rst_to_html(rst_doc["raw_data"]) readme_html = _rst_readme_to_html(readme_sha1) return readme_name, readme_url, readme_html diff --git a/swh/web/browse/views/content.py b/swh/web/browse/views/content.py index 7d0995a1..e9faf435 100644 --- a/swh/web/browse/views/content.py +++ b/swh/web/browse/views/content.py @@ -1,455 +1,461 @@ # Copyright (C) 2017-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import difflib from distutils.util import strtobool +from typing import Any, Dict, Optional -from django.http import HttpResponse, JsonResponse +from django.http import HttpRequest, HttpResponse, JsonResponse from django.shortcuts import redirect, render from swh.model.hashutil import hash_to_hex from swh.model.swhids import ObjectType from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import ( content_display_max_size, gen_link, prepare_content_for_display, request_content, ) from swh.web.common import archive, highlightjs, query from swh.web.common.exc import ( BadInputExc, NotFoundExc, http_status_code_message, sentry_capture_exception, ) 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/", + 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): +def content_raw(request: HttpRequest, query_string: str) -> HttpResponse: """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.*)/", + r"content/(?P.+)/diff/(?P.+)/", view_name="diff-contents", ) -def _contents_diff(request, from_query_string, to_query_string): +def _contents_diff( + request: HttpRequest, from_query_string: str, to_query_string: str +) -> HttpResponse: """ 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") + force_str = request.GET.get("force", "false") path = request.GET.get("path", None) language = "plaintext" - force = bool(strtobool(force)) + force = bool(strtobool(force_str)) if from_query_string == to_query_string: diff_str = "File renamed without changes" else: try: text_diff = True if from_query_string: content_from = request_content(from_query_string, max_size=None) content_from_display_data = prepare_content_for_display( content_from["raw_data"], content_from["mimetype"], path ) language = content_from_display_data["language"] content_from_size = content_from["length"] if not ( content_from["mimetype"].startswith("text/") or content_from["mimetype"] == "inode/x-empty" ): text_diff = False if text_diff and to_query_string: content_to = request_content(to_query_string, max_size=None) content_to_display_data = prepare_content_for_display( content_to["raw_data"], content_to["mimetype"], path ) language = content_to_display_data["language"] content_to_size = content_to["length"] if not ( content_to["mimetype"].startswith("text/") or content_to["mimetype"] == "inode/x-empty" ): text_diff = False diff_size = abs(content_to_size - content_from_size) if not text_diff: diff_str = "Diffs are not generated for non textual content" language = "plaintext" elif not force and diff_size > _auto_diff_size_limit: diff_str = "Large diffs are not automatically computed" language = "plaintext" else: if content_from: content_from_lines = ( content_from["raw_data"].decode("utf-8").splitlines(True) ) if content_from_lines and content_from_lines[-1][-1] != "\n": content_from_lines[-1] += "[swh-no-nl-marker]\n" if content_to: content_to_lines = ( content_to["raw_data"].decode("utf-8").splitlines(True) ) if content_to_lines and content_to_lines[-1][-1] != "\n": content_to_lines[-1] += "[swh-no-nl-marker]\n" diff_lines = difflib.unified_diff(content_from_lines, content_to_lines) diff_str = "".join(list(diff_lines)[2:]) except Exception as exc: sentry_capture_exception(exc) diff_str = str(exc) diff_data["diff_str"] = diff_str diff_data["language"] = language return JsonResponse(diff_data) -def _get_content_from_request(request): +def _get_content_from_request(request: HttpRequest) -> Dict[str, Any]: path = request.GET.get("path") if path is None: raise BadInputExc("The path query parameter must be provided.") snapshot = request.GET.get("snapshot") or request.GET.get("snapshot_id") origin_url = request.GET.get("origin_url") if snapshot is None and origin_url is None: raise BadInputExc( "The origin_url or snapshot query parameters must be provided." ) visit_id = int(request.GET.get("visit_id", 0)) snapshot_context = get_snapshot_context( snapshot_id=snapshot, origin_url=origin_url, path=path, timestamp=request.GET.get("timestamp"), visit_id=visit_id or None, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), browse_context="content", ) root_directory = snapshot_context["root_directory"] + assert root_directory is not None # to keep mypy happy return archive.lookup_directory_with_path(root_directory, path) @browse_route( - r"content/(?P[0-9a-z_:]*[0-9a-f]+.)/", + r"content/(?P[0-9a-z_:]*[0-9a-f]+)/", r"content/", view_name="browse-content", checksum_args=["query_string"], ) -def content_display(request, query_string=None): +def content_display( + request: HttpRequest, query_string: Optional[str] = None +) -> HttpResponse: """Django view that produces an HTML display of a content identified by its hash value. The URLs that points to it are :http:get:`/browse/content/[(algo_hash):](hash)/` :http:get:`/browse/content/` """ if query_string is None: # this case happens when redirected from origin/content or snapshot/content - content = _get_content_from_request(request) + content_data = _get_content_from_request(request) return redirect( reverse( "browse-content", - url_args={"query_string": f"sha1_git:{content['target']}"}, + url_args={"query_string": f"sha1_git:{content_data['target']}"}, query_params=request.GET, ), ) algo, checksum = query.parse_hash(query_string) checksum = hash_to_hex(checksum) origin_url = request.GET.get("origin_url") selected_language = request.GET.get("language") if not origin_url: origin_url = request.GET.get("origin") snapshot_id = request.GET.get("snapshot") or request.GET.get("snapshot_id") path = request.GET.get("path") content_data = {} - error_info = {"status_code": 200, "description": None} + error_info: Dict[str, Any] = {"status_code": 200, "description": None} try: content_data = request_content(query_string) except NotFoundExc as e: error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" snapshot_context = None if origin_url is not None or snapshot_id is not None: try: visit_id = int(request.GET.get("visit_id", 0)) snapshot_context = get_snapshot_context( origin_url=origin_url, snapshot_id=snapshot_id, timestamp=request.GET.get("timestamp"), visit_id=visit_id or None, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), path=path, browse_context="content", ) except NotFoundExc as e: - if str(e).startswith("Origin"): + if str(e).startswith("Origin") and origin_url is not None: 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}) + breadcrumbs.append({"name": filename, "url": ""}) - if path and root_dir != path: + if path and root_dir is not None and root_dir != path: dir_info = archive.lookup_directory_with_path(root_dir, path) directory_id = dir_info["target"] elif root_dir != path: directory_id = root_dir else: root_dir = None query_params = {"filename": filename} content_checksums = content_data.get("checksums", {}) content_url = reverse( "browse-content", url_args={"query_string": query_string}, ) content_raw_url = reverse( "browse-content-raw", url_args={"query_string": query_string}, query_params=query_params, ) content_metadata = ContentMetadata( object_type=ObjectType.CONTENT, object_id=content_checksums.get("sha1_git"), sha1=content_checksums.get("sha1"), sha1_git=content_checksums.get("sha1_git"), sha256=content_checksums.get("sha256"), blake2s256=content_checksums.get("blake2s256"), content_url=content_url, - mimetype=content_data.get("mimetype"), - encoding=content_data.get("encoding"), + mimetype=content_data.get("mimetype", ""), + encoding=content_data.get("encoding", ""), size=content_data.get("length", 0), - language=content_data.get("language"), + language=content_data.get("language", ""), root_directory=root_dir, path=f"/{path}" if path else None, filename=filename or "", directory=directory_id, revision=None, release=None, snapshot=None, origin_url=origin_url, ) swh_objects = [] if content_checksums: swh_objects.append( SWHObjectInfo( object_type=ObjectType.CONTENT, object_id=content_checksums.get("sha1_git"), ) ) if directory_id: swh_objects.append( SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=directory_id) ) if snapshot_context: if snapshot_context["revision_id"]: swh_objects.append( SWHObjectInfo( object_type=ObjectType.REVISION, object_id=snapshot_context["revision_id"], ) ) swh_objects.append( SWHObjectInfo( object_type=ObjectType.SNAPSHOT, object_id=snapshot_context["snapshot_id"], ) ) if snapshot_context["release_id"]: swh_objects.append( SWHObjectInfo( object_type=ObjectType.RELEASE, object_id=snapshot_context["release_id"], ) ) swhids_info = get_swhids_info( swh_objects, snapshot_context, extra_context=content_metadata, ) heading = "Content - %s" % content_checksums.get("sha1_git") if breadcrumbs: - content_path = "/".join([bc["name"] for bc in breadcrumbs]) + content_path = "/".join(bc["name"] for bc in breadcrumbs) heading += " - %s" % content_path return render( request, "browse/content.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"] if swhids_info else "", "swh_object_name": "Content", "swh_object_metadata": content_metadata, "content": content, "content_size": content_data.get("length"), "max_content_size": content_display_max_size, "filename": filename, "encoding": content_data.get("encoding"), "mimetype": mimetype, "language": language, "available_languages": available_languages, "breadcrumbs": breadcrumbs, "top_right_link": { "url": content_raw_url, "icon": swh_object_icons["content"], "text": "Raw File", }, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": True, "swhids_info": swhids_info, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], }, status=error_info["status_code"], ) diff --git a/swh/web/browse/views/directory.py b/swh/web/browse/views/directory.py index 09d852a1..124470e1 100644 --- a/swh/web/browse/views/directory.py +++ b/swh/web/browse/views/directory.py @@ -1,294 +1,304 @@ # Copyright (C) 2017-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import os +from typing import Any, Dict, Optional -from django.http import HttpResponse +from django.http import HttpRequest, HttpResponse from django.shortcuts import redirect, render from swh.model.swhids import ObjectType from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import gen_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, sentry_capture_exception, ) 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): +def _directory_browse( + request: HttpRequest, sha1_git: str, path: Optional[str] = None +) -> HttpResponse: root_sha1_git = sha1_git - error_info = {"status_code": 200, "description": None} + dir_sha1_git: Optional[str] = sha1_git + error_info: Dict[str, Any] = {"status_code": 200, "description": None} if path: try: dir_info = archive.lookup_directory_with_path(sha1_git, path) - sha1_git = dir_info["target"] + dir_sha1_git = dir_info["target"] except NotFoundExc as e: error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" - sha1_git = None + dir_sha1_git = None dirs, files = [], [] - if sha1_git is not None: - dirs, files = get_directory_entries(sha1_git) + if dir_sha1_git is not None: + dirs, files = get_directory_entries(dir_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"): + if str(e).startswith("Origin") and origin_url is not None: raw_dir_url = reverse( - "browse-directory", url_args={"sha1_git": sha1_git} + "browse-directory", url_args={"sha1_git": dir_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=ObjectType.DIRECTORY, - object_id=sha1_git, + object_id=dir_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}", + "directory_swhid": f"swh:1:dir:{dir_sha1_git}", "revision_context": False, "revision_swhid": None, } - swh_objects = [SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=sha1_git)] + swh_objects = [ + SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=dir_sha1_git) + ] if snapshot_context: if snapshot_context["revision_id"]: swh_objects.append( SWHObjectInfo( object_type=ObjectType.REVISION, object_id=snapshot_context["revision_id"], ) ) swh_objects.append( SWHObjectInfo( object_type=ObjectType.SNAPSHOT, object_id=snapshot_context["snapshot_id"], ) ) if snapshot_context["release_id"]: swh_objects.append( SWHObjectInfo( object_type=ObjectType.RELEASE, object_id=snapshot_context["release_id"], ) ) swhids_info = get_swhids_info(swh_objects, snapshot_context, dir_metadata) - heading = "Directory - %s" % sha1_git + heading = "Directory - %s" % dir_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"] and snapshot_context["revision_id"] is not None ): 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): +def directory_browse(request: HttpRequest, sha1_git: str) -> HttpResponse: """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): +def directory_browse_legacy( + request: HttpRequest, sha1_git: str, path: str +) -> HttpResponse: """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): +def _directory_resolve_content_path( + request: HttpRequest, sha1_git: str +) -> HttpResponse: """ 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")) + 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_capture_exception(exc) return HttpResponse(status=404) diff --git a/swh/web/browse/views/origin.py b/swh/web/browse/views/origin.py index 4995b27e..3f5ea8bc 100644 --- a/swh/web/browse/views/origin.py +++ b/swh/web/browse/views/origin.py @@ -1,310 +1,332 @@ -# Copyright (C) 2021 The Software Heritage developers +# Copyright (C) 2021-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information +from typing import Any, Dict, List, Optional, cast + +from django.http import HttpRequest, HttpResponse from django.shortcuts import redirect, render from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import ( browse_snapshot_directory, get_snapshot_context, ) from swh.web.common import archive from swh.web.common.exc import BadInputExc from swh.web.common.origin_visits import get_origin_visits from swh.web.common.utils import ( format_utc_iso_date, parse_iso8601_date_to_utc, redirect_to_new_route, reverse, ) @browse_route( r"origin/directory/", view_name="browse-origin-directory", ) -def origin_directory_browse(request): +def origin_directory_browse(request: HttpRequest) -> HttpResponse: """Django view for browsing the content of a directory associated to an origin for a given visit. The URL that points to it is :http:get:`/browse/origin/directory/` """ return browse_snapshot_directory( request, origin_url=request.GET.get("origin_url"), snapshot_id=request.GET.get("snapshot"), timestamp=request.GET.get("timestamp"), path=request.GET.get("path"), ) @browse_route( r"origin/(?P.+)/visit/(?P.+)/directory/", r"origin/(?P.+)/visit/(?P.+)/directory/(?P.+)/", r"origin/(?P.+)/directory/(?P.+)/", r"origin/(?P.+)/directory/", view_name="browse-origin-directory-legacy", ) -def origin_directory_browse_legacy(request, origin_url, timestamp=None, path=None): +def origin_directory_browse_legacy( + request: HttpRequest, + origin_url: str, + timestamp: Optional[str] = None, + path: Optional[str] = None, +) -> HttpResponse: """Django view for browsing the content of a directory associated to an origin for a given visit. The URLs that point to it are :http:get:`/browse/origin/(origin_url)/directory/[(path)/]` and :http:get:`/browse/origin/(origin_url)/visit/(timestamp)/directory/[(path)/]` """ return browse_snapshot_directory( request, origin_url=origin_url, snapshot_id=request.GET.get("snapshot"), timestamp=timestamp, path=path, ) @browse_route( r"origin/content/", view_name="browse-origin-content", ) -def origin_content_browse(request): +def origin_content_browse(request: HttpRequest) -> HttpResponse: """ This route is deprecated; use http:get:`/browse/content` instead Django view that produces an HTML display of a content associated to an origin for a given visit. The URL that points to it is :http:get:`/browse/origin/content/` """ return redirect_to_new_route(request, "browse-content") @browse_route( r"origin/(?P.+)/visit/(?P.+)/content/(?P.+)/", r"origin/(?P.+)/content/(?P.+)/", r"origin/(?P.+)/content/", view_name="browse-origin-content-legacy", ) -def origin_content_browse_legacy(request, origin_url, path=None, timestamp=None): +def origin_content_browse_legacy( + request: HttpRequest, + origin_url: str, + path: Optional[str] = None, + timestamp: Optional[str] = None, +) -> HttpResponse: """ This route is deprecated; use http:get:`/browse/content` instead Django view that produces an HTML display of a content associated to an origin for a given visit. The URLs that point to it are :http:get:`/browse/origin/(origin_url)/content/(path)/` and :http:get:`/browse/origin/(origin_url)/visit/(timestamp)/content/(path)/` """ return redirect_to_new_route(request, "browse-content") @browse_route( r"origin/log/", view_name="browse-origin-log", ) -def origin_log_browse(request): +def origin_log_browse(request: HttpRequest) -> HttpResponse: """ This route is deprecated; use http:get:`/browse/snapshot/log` instead Django view that produces an HTML display of revisions history (aka the commit log) associated to a software origin. The URL that points to it is :http:get:`/browse/origin/log/` """ return redirect_to_new_route(request, "browse-snapshot-log") @browse_route( r"origin/(?P.+)/visit/(?P.+)/log/", r"origin/(?P.+)/log/", view_name="browse-origin-log-legacy", ) -def origin_log_browse_legacy(request, origin_url, timestamp=None): +def origin_log_browse_legacy( + request: HttpRequest, origin_url: str, timestamp: Optional[str] = None +) -> HttpResponse: """ This route is deprecated; use http:get:`/browse/snapshot/log` instead Django view that produces an HTML display of revisions history (aka the commit log) associated to a software origin. The URLs that point to it are :http:get:`/browse/origin/(origin_url)/log/` and :http:get:`/browse/origin/(origin_url)/visit/(timestamp)/log/` """ return redirect_to_new_route( request, "browse-snapshot-log", ) @browse_route( r"origin/branches/", view_name="browse-origin-branches", ) -def origin_branches_browse(request): +def origin_branches_browse(request: HttpRequest) -> HttpResponse: """ This route is deprecated; use http:get:`/browse/snapshot/branches` instead Django view that produces an HTML display of the list of branches associated to an origin for a given visit. The URL that points to it is :http:get:`/browse/origin/branches/` """ return redirect_to_new_route(request, "browse-snapshot-branches") @browse_route( r"origin/(?P.+)/visit/(?P.+)/branches/", r"origin/(?P.+)/branches/", view_name="browse-origin-branches-legacy", ) -def origin_branches_browse_legacy(request, origin_url, timestamp=None): +def origin_branches_browse_legacy( + request: HttpRequest, origin_url: str, timestamp: Optional[str] = None +) -> HttpResponse: """ This route is deprecated; use http:get:`/browse/snapshot/branches` instead Django view that produces an HTML display of the list of branches associated to an origin for a given visit. The URLs that point to it are :http:get:`/browse/origin/(origin_url)/branches/` and :http:get:`/browse/origin/(origin_url)/visit/(timestamp)/branches/` """ return redirect_to_new_route(request, "browse-snapshot-branches") @browse_route( r"origin/releases/", view_name="browse-origin-releases", ) -def origin_releases_browse(request): +def origin_releases_browse(request: HttpRequest) -> HttpResponse: """ This route is deprecated; use http:get:`/browse/snapshot/releases` instead Django view that produces an HTML display of the list of releases associated to an origin for a given visit. The URL that points to it is :http:get:`/browse/origin/releases/` """ return redirect_to_new_route(request, "browse-snapshot-releases") @browse_route( r"origin/(?P.+)/visit/(?P.+)/releases/", r"origin/(?P.+)/releases/", view_name="browse-origin-releases-legacy", ) -def origin_releases_browse_legacy(request, origin_url, timestamp=None): +def origin_releases_browse_legacy( + request: HttpRequest, origin_url: str, timestamp: Optional[str] = None +) -> HttpResponse: """ This route is deprecated; use http:get:`/browse/snapshot/releases` instead Django view that produces an HTML display of the list of releases associated to an origin for a given visit. The URLs that point to it are :http:get:`/browse/origin/(origin_url)/releases/` and :http:get:`/browse/origin/(origin_url)/visit/(timestamp)/releases/` """ return redirect_to_new_route(request, "browse-snapshot-releases") -def _origin_visits_browse(request, origin_url): +def _origin_visits_browse( + request: HttpRequest, origin_url: Optional[str] +) -> HttpResponse: if origin_url is None: raise BadInputExc("An origin URL must be provided as query parameter.") origin_info = archive.lookup_origin({"url": origin_url}) - origin_visits = get_origin_visits(origin_info) + origin_visits = cast(List[Dict[str, Any]], get_origin_visits(origin_info)) + snapshot_context = get_snapshot_context(origin_url=origin_url) for i, visit in enumerate(origin_visits): url_date = format_utc_iso_date(visit["date"], "%Y-%m-%dT%H:%M:%SZ") visit["formatted_date"] = format_utc_iso_date(visit["date"]) query_params = {"origin_url": origin_url, "timestamp": url_date} if i < len(origin_visits) - 1: if visit["date"] == origin_visits[i + 1]["date"]: query_params = {"visit_id": visit["visit"]} if i > 0: if visit["date"] == origin_visits[i - 1]["date"]: query_params = {"visit_id": visit["visit"]} snapshot = visit["snapshot"] if visit["snapshot"] else "" visit["url"] = reverse( "browse-origin-directory", query_params=query_params, ) if not snapshot: visit["snapshot"] = "" visit["date"] = parse_iso8601_date_to_utc(visit["date"]).timestamp() heading = "Origin visits - %s" % origin_url return render( request, "browse/origin-visits.html", { "heading": heading, "swh_object_name": "Visits", "swh_object_metadata": origin_info, "origin_visits": origin_visits, "origin_info": origin_info, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": False, }, ) @browse_route(r"origin/visits/", view_name="browse-origin-visits") -def origin_visits_browse(request): +def origin_visits_browse(request: HttpRequest) -> HttpResponse: """Django view that produces an HTML display of visits reporting for a given origin. The URL that points to it is :http:get:`/browse/origin/visits/`. """ return _origin_visits_browse(request, request.GET.get("origin_url")) @browse_route( r"origin/(?P.+)/visits/", view_name="browse-origin-visits-legacy" ) -def origin_visits_browse_legacy(request, origin_url): +def origin_visits_browse_legacy(request: HttpRequest, origin_url: str) -> HttpResponse: """Django view that produces an HTML display of visits reporting for a given origin. The URL that points to it is :http:get:`/browse/origin/(origin_url)/visits/`. """ return _origin_visits_browse(request, origin_url) @browse_route(r"origin/", view_name="browse-origin") -def origin_browse(request): +def origin_browse(request: HttpRequest) -> HttpResponse: """Django view that redirects to the display of the latest archived snapshot for a given software origin. """ last_snapshot_url = reverse( "browse-origin-directory", query_params=request.GET, ) return redirect(last_snapshot_url) @browse_route(r"origin/(?P.+)/", view_name="browse-origin-legacy") -def origin_browse_legacy(request, origin_url): +def origin_browse_legacy(request: HttpRequest, origin_url: str) -> HttpResponse: """Django view that redirects to the display of the latest archived snapshot for a given software origin. """ last_snapshot_url = reverse( "browse-origin-directory", query_params={"origin_url": origin_url, **request.GET}, ) return redirect(last_snapshot_url) diff --git a/swh/web/browse/views/release.py b/swh/web/browse/views/release.py index 6f3d9dec..16670106 100644 --- a/swh/web/browse/views/release.py +++ b/swh/web/browse/views/release.py @@ -1,243 +1,247 @@ # Copyright (C) 2017-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information +from typing import Optional + +from django.http import HttpRequest, HttpResponse from django.shortcuts import render from swh.model.swhids import ObjectType from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import ( gen_content_link, gen_directory_link, gen_link, gen_person_mail_link, gen_release_link, gen_revision_link, ) from swh.web.common import archive from swh.web.common.exc import NotFoundExc, sentry_capture_exception from swh.web.common.identifiers import get_swhids_info -from swh.web.common.typing import ReleaseMetadata, SWHObjectInfo +from swh.web.common.typing import ReleaseMetadata, SnapshotContext, 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): +def release_browse(request: HttpRequest, sha1_git: str) -> HttpResponse: """ 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 = {} + snapshot_context: Optional[SnapshotContext] = None 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 = int(request.GET.get("visit_id", 0)) if origin_url: try: snapshot_context = get_snapshot_context( snapshot_id, origin_url, timestamp, visit_id or None, release_name=release["name"], ) except NotFoundExc as e: raw_rel_url = reverse("browse-release", url_args={"sha1_git": sha1_git}) error_message = ( "The Software Heritage archive has a release " "with the hash you provided but the origin " "mentioned in your request appears broken: %s. " "Please check the URL and try again.\n\n" "Nevertheless, you can still browse the release " "without origin information: %s" % (gen_link(origin_url), gen_link(raw_rel_url)) ) if str(e).startswith("Origin"): raise NotFoundExc(error_message) else: raise e origin_info = snapshot_context["origin_info"] elif snapshot_id: snapshot_context = get_snapshot_context( snapshot_id, release_name=release["name"] ) - snapshot_id = snapshot_context.get("snapshot_id", None) + if snapshot_context is not None: + snapshot_id = snapshot_context.get("snapshot_id", None) release_metadata = ReleaseMetadata( object_type=ObjectType.RELEASE, object_id=sha1_git, release=sha1_git, author=release["author"]["fullname"] if release["author"] else "None", author_url=gen_person_mail_link(release["author"]) if release["author"] else "None", date=format_utc_iso_date(release["date"]), name=release["name"], synthetic=release["synthetic"], target=release["target"], target_type=release["target_type"], snapshot=snapshot_id, origin_url=origin_url, ) release_note_lines = [] if release["message"]: release_note_lines = release["message"].split("\n") swh_objects = [SWHObjectInfo(object_type=ObjectType.RELEASE, object_id=sha1_git)] vault_cooking = None rev_directory = None target_link = None if release["target_type"] == ObjectType.REVISION.name.lower(): target_link = gen_revision_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) try: revision = archive.lookup_revision(release["target"]) rev_directory = revision["directory"] vault_cooking = { "directory_context": True, "directory_swhid": f"swh:1:dir:{rev_directory}", "revision_context": True, "revision_swhid": f"swh:1:rev:{release['target']}", } swh_objects.append( SWHObjectInfo( object_type=ObjectType.REVISION, object_id=release["target"] ) ) swh_objects.append( SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=rev_directory) ) except Exception as exc: sentry_capture_exception(exc) elif release["target_type"] == ObjectType.DIRECTORY.name.lower(): target_link = gen_directory_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) try: # check directory exists archive.lookup_directory(release["target"]) vault_cooking = { "directory_context": True, "directory_swhid": f"swh:1:dir:{release['target']}", "revision_context": False, "revision_swhid": None, } swh_objects.append( SWHObjectInfo( object_type=ObjectType.DIRECTORY, object_id=release["target"] ) ) except Exception as exc: sentry_capture_exception(exc) elif release["target_type"] == ObjectType.CONTENT.name.lower(): target_link = gen_content_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) swh_objects.append( SWHObjectInfo(object_type=ObjectType.CONTENT, object_id=release["target"]) ) elif release["target_type"] == ObjectType.RELEASE.name.lower(): target_link = gen_release_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) rev_directory_url = None if rev_directory is not None: if origin_info: rev_directory_url = reverse( "browse-origin-directory", query_params={ "origin_url": origin_info["url"], "release": release["name"], "snapshot": snapshot_id, }, ) elif snapshot_id: rev_directory_url = reverse( "browse-snapshot-directory", url_args={"snapshot_id": snapshot_id}, query_params={"release": release["name"]}, ) else: rev_directory_url = reverse( "browse-directory", url_args={"sha1_git": rev_directory} ) directory_link = None if rev_directory_url is not None: directory_link = gen_link(rev_directory_url, rev_directory) release["directory_link"] = directory_link release["target_link"] = target_link if snapshot_context: snapshot_id = snapshot_context["snapshot_id"] if snapshot_id: swh_objects.append( SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id) ) swhids_info = get_swhids_info(swh_objects, snapshot_context) note_header = "None" if len(release_note_lines) > 0: note_header = release_note_lines[0] release["note_header"] = note_header release["note_body"] = "\n".join(release_note_lines[1:]) heading = "Release - %s" % release["name"] if snapshot_context: context_found = "snapshot: %s" % snapshot_context["snapshot_id"] if origin_info: context_found = "origin: %s" % origin_info["url"] heading += " - %s" % context_found return render( request, "browse/release.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"], "swh_object_name": "Release", "swh_object_metadata": release_metadata, "release": release, "snapshot_context": snapshot_context, "show_actions": True, "breadcrumbs": None, "vault_cooking": vault_cooking, "top_right_link": None, "swhids_info": swhids_info, }, ) diff --git a/swh/web/browse/views/revision.py b/swh/web/browse/views/revision.py index b3f020ed..4174b191 100644 --- a/swh/web/browse/views/revision.py +++ b/swh/web/browse/views/revision.py @@ -1,589 +1,600 @@ # Copyright (C) 2017-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import hashlib import json import textwrap +from typing import Any, Dict, List, Optional -from django.http import JsonResponse +from django.http import HttpRequest, HttpResponse, JsonResponse from django.shortcuts import render from django.utils.safestring import mark_safe from swh.model.hashutil import hash_to_bytes from swh.model.swhids import CoreSWHID, ObjectType from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import ( content_display_max_size, format_log_entries, gen_link, gen_person_mail_link, gen_revision_url, get_directory_entries, get_readme_to_display, get_revision_log_url, prepare_content_for_display, request_content, ) from swh.web.common import archive from swh.web.common.exc import NotFoundExc, http_status_code_message from swh.web.common.identifiers import get_swhids_info -from swh.web.common.typing import RevisionMetadata, SWHObjectInfo +from swh.web.common.typing import RevisionMetadata, SnapshotContext, 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): +def _gen_content_url( + revision: Dict[str, Any], + query_string: str, + path: str, + snapshot_context: Optional[SnapshotContext], +) -> str: 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): +def _gen_diff_link(idx: int, diff_anchor: str, link_text: str) -> str: 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): +def _gen_revision_changes_list( + revision: Dict[str, Any], + changes: List[Dict[str, Any]], + snapshot_context: Optional[SnapshotContext], +) -> str: """ 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): +def _revision_diff(request: HttpRequest, sha1_git: str) -> HttpResponse: """ 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) + visit_id_str = request.GET.get("visit_id", None) + visit_id = int(visit_id_str) if visit_id_str is not None else 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): +def revision_log_browse(request: HttpRequest, sha1_git: str) -> HttpResponse: """ Django view that produces an HTML display of the history log for a revision identified by its id. The url that points to it is :http:get:`/browse/revision/(sha1_git)/log/` """ origin_url = request.GET.get("origin_url") snapshot_id = request.GET.get("snapshot") snapshot_context = None if origin_url or snapshot_id: visit_id = int(request.GET.get("visit_id", 0)) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=request.GET.get("timestamp"), visit_id=visit_id or None, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=sha1_git, ) per_page = int(request.GET.get("per_page", NB_LOG_ENTRIES)) offset = int(request.GET.get("offset", 0)) revs_ordering = request.GET.get("revs_ordering", "committer_date") session_key = "rev_%s_log_ordering_%s" % (sha1_git, revs_ordering) rev_log_session = request.session.get(session_key, None) rev_log = [] revs_walker_state = None if rev_log_session: rev_log = rev_log_session["rev_log"] revs_walker_state = rev_log_session["revs_walker_state"] if len(rev_log) < offset + per_page: revs_walker = archive.get_revisions_walker( revs_ordering, sha1_git, max_revs=offset + per_page + 1, state=revs_walker_state, ) rev_log += [rev["id"] for rev in revs_walker] revs_walker_state = revs_walker.export_state() revs = rev_log[offset : offset + per_page] revision_log = archive.lookup_revision_multiple(revs) request.session[session_key] = { "rev_log": rev_log, "revs_walker_state": revs_walker_state, } revs_ordering = request.GET.get("revs_ordering", "") prev_log_url = None if len(rev_log) > offset + per_page: prev_log_url = reverse( "browse-revision-log", url_args={"sha1_git": sha1_git}, query_params={ "per_page": per_page, "offset": offset + per_page, "revs_ordering": revs_ordering or None, }, ) next_log_url = None if offset != 0: next_log_url = reverse( "browse-revision-log", url_args={"sha1_git": sha1_git}, query_params={ "per_page": per_page, "offset": offset - per_page, "revs_ordering": revs_ordering or None, }, ) revision_log_data = format_log_entries(revision_log, per_page) swh_rev_id = str( CoreSWHID(object_type=ObjectType.REVISION, object_id=hash_to_bytes(sha1_git)) ) return render( request, "browse/revision-log.html", { "heading": "Revision history", "swh_object_id": swh_rev_id, "swh_object_name": "Revisions history", "swh_object_metadata": None, "revision_log": revision_log_data, "revs_ordering": revs_ordering, "next_log_url": next_log_url, "prev_log_url": prev_log_url, "breadcrumbs": None, "top_right_link": None, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": True, "swhids_info": None, }, ) @browse_route( r"revision/(?P[0-9a-f]+)/", view_name="browse-revision", checksum_args=["sha1_git"], ) -def revision_browse(request, sha1_git): +def revision_browse(request: HttpRequest, sha1_git: str) -> HttpResponse: """ 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 = int(request.GET.get("visit_id", 0)) snapshot_id = request.GET.get("snapshot_id") if not snapshot_id: snapshot_id = request.GET.get("snapshot") path = request.GET.get("path") dir_id = None dirs, files = [], [] content_data = {} if origin_url: try: snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=visit_id or None, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=sha1_git, path=path, ) except NotFoundExc as e: raw_rev_url = reverse("browse-revision", url_args={"sha1_git": sha1_git}) error_message = ( "The Software Heritage archive has a revision " "with the hash you provided but the origin " "mentioned in your request appears broken: %s. " "Please check the URL and try again.\n\n" "Nevertheless, you can still browse the revision " "without origin information: %s" % (gen_link(origin_url), gen_link(raw_rev_url)) ) if str(e).startswith("Origin"): raise NotFoundExc(error_message) else: raise e origin_info = snapshot_context["origin_info"] snapshot_id = snapshot_context["snapshot_id"] elif snapshot_id: snapshot_context = get_snapshot_context(snapshot_id) - error_info = {"status_code": 200, "description": None} + error_info: Dict[str, Any] = {"status_code": 200, "description": None} if path: try: file_info = archive.lookup_directory_with_path(revision["directory"], path) if file_info["type"] == "dir": dir_id = file_info["target"] else: query_string = "sha1_git:" + file_info["target"] content_data = request_content(query_string) except NotFoundExc as e: error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" else: dir_id = revision["directory"] if dir_id: path = "" if path is None else (path + "/") dirs, files = get_directory_entries(dir_id) revision_metadata = RevisionMetadata( object_type=ObjectType.REVISION, object_id=sha1_git, revision=sha1_git, author=revision["author"]["fullname"] if revision["author"] else "None", author_url=gen_person_mail_link(revision["author"]) if revision["author"] else "None", committer=revision["committer"]["fullname"] if revision["committer"] else "None", committer_url=gen_person_mail_link(revision["committer"]) if revision["committer"] else "None", committer_date=format_utc_iso_date(revision["committer_date"]), date=format_utc_iso_date(revision["date"]), directory=revision["directory"], merge=revision["merge"], metadata=json.dumps( revision["metadata"], sort_keys=True, indent=4, separators=(",", ": ") ), parents=revision["parents"], synthetic=revision["synthetic"], type=revision["type"], snapshot=snapshot_id, origin_url=origin_url, ) message_lines = ["None"] if revision["message"]: message_lines = revision["message"].split("\n") parents = [] for p in revision["parents"]: parent_url = gen_revision_url(p, snapshot_context) parents.append({"id": p, "url": parent_url}) path_info = gen_path_info(path) query_params = snapshot_context["query_params"] if snapshot_context else {} breadcrumbs = [] breadcrumbs.append( { "name": revision["directory"][:7], "url": reverse( "browse-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ), } ) for pi in path_info: query_params["path"] = pi["path"] breadcrumbs.append( { "name": pi["name"], "url": reverse( "browse-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ), } ) vault_cooking = { "directory_context": False, "directory_swhid": None, "revision_context": True, "revision_swhid": f"swh:1:rev:{sha1_git}", } swh_objects = [SWHObjectInfo(object_type=ObjectType.REVISION, object_id=sha1_git)] content = None content_size = None filename = None mimetype = None language = None readme_name = None readme_url = None readme_html = None readmes = {} extra_context = dict(revision_metadata) extra_context["path"] = f"/{path}" if path else None if content_data: breadcrumbs[-1]["url"] = None content_size = content_data["length"] mimetype = content_data["mimetype"] if content_data["raw_data"]: content_display_data = prepare_content_for_display( content_data["raw_data"], content_data["mimetype"], path ) content = content_display_data["content_data"] language = content_display_data["language"] mimetype = content_display_data["mimetype"] if path: filename = path_info[-1]["name"] query_params["filename"] = filename filepath = "/".join(pi["name"] for pi in path_info[:-1]) extra_context["path"] = f"/{filepath}/" if filepath else "/" extra_context["filename"] = filename top_right_link = { "url": reverse( "browse-content-raw", url_args={"query_string": query_string}, query_params={"filename": filename}, ), "icon": swh_object_icons["content"], "text": "Raw File", } swh_objects.append( SWHObjectInfo(object_type=ObjectType.CONTENT, object_id=file_info["target"]) ) else: for d in dirs: if d["type"] == "rev": d["url"] = reverse( "browse-revision", url_args={"sha1_git": d["target"]} ) else: query_params["path"] = path + d["name"] d["url"] = reverse( "browse-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ) for f in files: query_params["path"] = path + f["name"] f["url"] = reverse( "browse-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ) if f["name"].lower().startswith("readme"): readmes[f["name"]] = f["checksums"]["sha1"] readme_name, readme_url, readme_html = get_readme_to_display(readmes) top_right_link = { "url": get_revision_log_url(sha1_git, snapshot_context), "icon": swh_object_icons["revisions history"], "text": "History", } vault_cooking["directory_context"] = True vault_cooking["directory_swhid"] = f"swh:1:dir:{dir_id}" swh_objects.append( SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=dir_id) ) query_params.pop("path", None) diff_revision_url = reverse( "diff-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ) if snapshot_id: swh_objects.append( SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id) ) swhids_info = get_swhids_info(swh_objects, snapshot_context, extra_context) heading = "Revision - %s - %s" % ( sha1_git[:7], textwrap.shorten(message_lines[0], width=70), ) if snapshot_context: context_found = "snapshot: %s" % snapshot_context["snapshot_id"] if origin_info: context_found = "origin: %s" % origin_info["url"] heading += " - %s" % context_found return render( request, "browse/revision.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"], "swh_object_name": "Revision", "swh_object_metadata": revision_metadata, "message_header": message_lines[0], "message_body": "\n".join(message_lines[1:]), "parents": parents, "snapshot_context": snapshot_context, "dirs": dirs, "files": files, "content": content, "content_size": content_size, "max_content_size": content_display_max_size, "filename": filename, "encoding": content_data.get("encoding"), "mimetype": mimetype, "language": language, "readme_name": readme_name, "readme_url": readme_url, "readme_html": readme_html, "breadcrumbs": breadcrumbs, "top_right_link": top_right_link, "vault_cooking": vault_cooking, "diff_revision_url": diff_revision_url, "show_actions": True, "swhids_info": swhids_info, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], }, status=error_info["status_code"], ) diff --git a/swh/web/browse/views/snapshot.py b/swh/web/browse/views/snapshot.py index 2956d830..4bc775b5 100644 --- a/swh/web/browse/views/snapshot.py +++ b/swh/web/browse/views/snapshot.py @@ -1,225 +1,237 @@ -# Copyright (C) 2018-2019 The Software Heritage developers +# Copyright (C) 2018-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information +from typing import Optional +from django.http import HttpRequest, HttpResponse from django.shortcuts import redirect from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import ( browse_snapshot_branches, browse_snapshot_directory, browse_snapshot_log, browse_snapshot_releases, get_snapshot_context, ) from swh.web.common.exc import BadInputExc from swh.web.common.utils import redirect_to_new_route, reverse -def get_snapshot_from_request(request): +def get_snapshot_from_request(request: HttpRequest) -> str: snapshot = request.GET.get("snapshot") if snapshot: return snapshot if request.GET.get("origin_url") is None: raise BadInputExc("An origin URL must be provided as a query parameter.") return get_snapshot_context( origin_url=request.GET.get("origin_url"), timestamp=request.GET.get("timestamp") )["snapshot_id"] @browse_route( r"snapshot/(?P[0-9a-f]+)/", view_name="browse-snapshot", checksum_args=["snapshot_id"], ) -def snapshot_browse(request, snapshot_id): +def snapshot_browse(request: HttpRequest, snapshot_id: str) -> HttpResponse: """Django view for browsing the content of a snapshot. The url that points to it is :http:get:`/browse/snapshot/(snapshot_id)/` """ browse_snapshot_url = reverse( "browse-snapshot-directory", url_args={"snapshot_id": snapshot_id}, query_params=request.GET, ) return redirect(browse_snapshot_url) @browse_route( r"snapshot/(?P[0-9a-f]+)/directory/", view_name="browse-snapshot-directory", checksum_args=["snapshot_id"], ) -def snapshot_directory_browse(request, snapshot_id): +def snapshot_directory_browse(request: HttpRequest, snapshot_id: str) -> HttpResponse: """Django view for browsing the content of a directory collected in a snapshot. The URL that points to it is :http:get:`/browse/snapshot/(snapshot_id)/directory/` """ return browse_snapshot_directory( request, snapshot_id=snapshot_id, path=request.GET.get("path"), origin_url=request.GET.get("origin_url"), ) @browse_route( r"snapshot/(?P[0-9a-f]+)/directory/(?P.+)/", view_name="browse-snapshot-directory-legacy", checksum_args=["snapshot_id"], ) -def snapshot_directory_browse_legacy(request, snapshot_id, path=None): +def snapshot_directory_browse_legacy( + request: HttpRequest, snapshot_id: str, path: Optional[str] = None +) -> HttpResponse: """Django view for browsing the content of a directory collected in a snapshot. The URL that points to it is :http:get:`/browse/snapshot/(snapshot_id)/directory/(path)/` """ origin_url = request.GET.get("origin_url", None) if not origin_url: origin_url = request.GET.get("origin", None) return browse_snapshot_directory( request, snapshot_id=snapshot_id, path=path, origin_url=origin_url ) @browse_route( r"snapshot/(?P[0-9a-f]+)/content/", view_name="browse-snapshot-content", checksum_args=["snapshot_id"], ) -def snapshot_content_browse(request, snapshot_id): +def snapshot_content_browse(request: HttpRequest, snapshot_id: str) -> HttpResponse: """ This route is deprecated; use http:get:`/browse/content` instead Django view that produces an HTML display of a content collected in a snapshot. The url that points to it is :http:get:`/browse/snapshot/(snapshot_id)/content/` """ return redirect_to_new_route(request, "browse-content") @browse_route( r"snapshot/(?P[0-9a-f]+)/content/(?P.+)/", view_name="browse-snapshot-content-legacy", checksum_args=["snapshot_id"], ) -def snapshot_content_browse_legacy(request, snapshot_id, path): +def snapshot_content_browse_legacy( + request: HttpRequest, snapshot_id: str, path: str +) -> HttpResponse: """ This route is deprecated; use http:get:`/browse/content` instead Django view that produces an HTML display of a content collected in a snapshot. The url that points to it is :http:get:`/browse/snapshot/(snapshot_id)/content/(path)/` """ return redirect_to_new_route(request, "browse-content") @browse_route( r"snapshot/(?P[0-9a-f]+)/log/", r"snapshot/log/", view_name="browse-snapshot-log", checksum_args=["snapshot_id"], ) -def snapshot_log_browse(request, snapshot_id=None): +def snapshot_log_browse( + request: HttpRequest, snapshot_id: Optional[str] = None +) -> HttpResponse: """Django view that produces an HTML display of revisions history (aka the commit log) collected in a snapshot. The URLs that point to it are :http:get:`/browse/snapshot/(snapshot_id)/log/` and :http:get:`/browse/snapshot/log/` """ if snapshot_id is None: # This case happens when redirected from /origin/log snapshot_id = get_snapshot_from_request(request) # Redirect to the same route with snapshot_id return redirect( reverse( "browse-snapshot-log", url_args={"snapshot_id": snapshot_id}, query_params=request.GET, ), ) return browse_snapshot_log( request, snapshot_id=snapshot_id, origin_url=request.GET.get("origin_url"), timestamp=request.GET.get("timestamp"), ) @browse_route( r"snapshot/(?P[0-9a-f]+)/branches/", r"snapshot/branches/", view_name="browse-snapshot-branches", checksum_args=["snapshot_id"], ) -def snapshot_branches_browse(request, snapshot_id=None): +def snapshot_branches_browse( + request: HttpRequest, snapshot_id: Optional[str] = None +) -> HttpResponse: """Django view that produces an HTML display of the list of branches collected in a snapshot. The URLs that point to it are :http:get:`/browse/snapshot/(snapshot_id)/branches/` and :http:get:`/browse/snapshot/branches/` """ if snapshot_id is None: # This case happens when redirected from /origin/branches snapshot_id = get_snapshot_from_request(request) # Redirect to the same route with the newest snapshot_id # for the given origin return redirect( reverse( "browse-snapshot-branches", url_args={"snapshot_id": snapshot_id}, query_params=request.GET, ), ) return browse_snapshot_branches( request, snapshot_id=snapshot_id, origin_url=request.GET.get("origin_url"), timestamp=request.GET.get("timestamp"), branch_name_include=request.GET.get("name_include"), ) @browse_route( r"snapshot/(?P[0-9a-f]+)/releases/", r"snapshot/releases/", view_name="browse-snapshot-releases", checksum_args=["snapshot_id"], ) -def snapshot_releases_browse(request, snapshot_id=None): +def snapshot_releases_browse( + request: HttpRequest, snapshot_id: Optional[str] = None +) -> HttpResponse: """Django view that produces an HTML display of the list of releases collected in a snapshot. The URLs that point to it are :http:get:`/browse/snapshot/(snapshot_id)/releases/` :http:get:`/browse/snapshot/releases/` """ if snapshot_id is None: # This case happens when redirected from /origin/releases snapshot_id = get_snapshot_from_request(request) # Redirect to the same route with the newest snapshot_id # for the given origin return redirect( reverse( "browse-snapshot-releases", url_args={"snapshot_id": snapshot_id}, query_params=request.GET, ), ) return browse_snapshot_releases( request, snapshot_id=snapshot_id, origin_url=request.GET.get("origin_url"), timestamp=request.GET.get("timestamp"), release_name_include=request.GET.get("name_include"), ) diff --git a/swh/web/common/identifiers.py b/swh/web/common/identifiers.py index d7160d22..226e63e6 100644 --- a/swh/web/common/identifiers.py +++ b/swh/web/common/identifiers.py @@ -1,387 +1,387 @@ -# Copyright (C) 2020-2021 The Software Heritage developers +# Copyright (C) 2020-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information -from typing import Any, Dict, Iterable, List, Optional +from typing import Any, Dict, Iterable, List, Mapping, 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.swhids 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: 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_id = hash_to_bytes(object_id) obj_swhid = str( QualifiedSWHID( 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] = str(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"] == 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"] = str(query_dict["path"]).lstrip("/") elif object_type == ObjectType.DIRECTORY: object_id = directory # remove leading and trailing slashes from SWHID directory path query_dict["path"] = str(query_dict["path"]).strip("/") # 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[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[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].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, + extra_context: Optional[Mapping[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 != ObjectType.SNAPSHOT: swhid_context["visit"] = gen_swhid( ObjectType.SNAPSHOT, snapshot_context["snapshot_id"] ) if object_type in (ObjectType.CONTENT, ObjectType.DIRECTORY): if snapshot_context["release_id"] is not None: swhid_context["anchor"] = gen_swhid( ObjectType.RELEASE, snapshot_context["release_id"] ) elif snapshot_context["revision_id"] is not None: swhid_context["anchor"] = gen_swhid( ObjectType.REVISION, snapshot_context["revision_id"] ) 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( 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 != ObjectType.DIRECTORY or extra_context["root_directory"] != object_id ) ): swhid_context["anchor"] = gen_swhid( 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 == ObjectType.CONTENT: path += extra_context["filename"] 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/typing.py b/swh/web/common/typing.py index 6791b78a..6a02ad3e 100644 --- a/swh/web/common/typing.py +++ b/swh/web/common/typing.py @@ -1,263 +1,263 @@ # Copyright (C) 2020-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from 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.swhids 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 associated 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""" directory_url: Optional[str] """optional root directory URL associated to the snapshot""" class SWHObjectInfo(TypedDict): object_type: ObjectType - object_id: str + object_id: Optional[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 + directory: Optional[str] + nb_files: Optional[int] + nb_dirs: Optional[int] + sum_file_sizes: Optional[int] root_directory: Optional[str] - path: str + path: Optional[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, total=False): 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""" note: Optional[str] """Optional note associated to the request, for instance rejection reason""" 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)"""