diff --git a/swh/web/browse/snapshot_context.py b/swh/web/browse/snapshot_context.py index be48be0f..a11fdf67 100644 --- a/swh/web/browse/snapshot_context.py +++ b/swh/web/browse/snapshot_context.py @@ -1,1460 +1,1459 @@ # Copyright (C) 2018-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information # Utility module for browsing the archive in a snapshot context. from collections import defaultdict from copy import copy from typing import Any, Dict, List, Optional, Tuple from django.core.cache import cache from django.shortcuts import render from django.template.defaultfilters import filesizeformat from django.utils.html import escape from swh.model.identifiers import ( CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT, snapshot_identifier, swhid, ) from swh.web.browse.utils import ( content_display_max_size, format_log_entries, gen_content_link, gen_directory_link, gen_release_link, gen_revision_link, gen_revision_log_link, gen_revision_url, gen_snapshot_link, get_directory_entries, get_readme_to_display, prepare_content_for_display, request_content, ) from swh.web.common import highlightjs, service from swh.web.common.exc import BadInputExc, NotFoundExc, handle_view_exception from swh.web.common.identifiers import get_swhids_info from swh.web.common.origin_visits import get_origin_visit from swh.web.common.typing import ( ContentMetadata, DirectoryMetadata, OriginInfo, SnapshotBranchInfo, SnapshotContext, SnapshotReleaseInfo, SWHObjectInfo, ) from swh.web.common.utils import ( format_utc_iso_date, gen_path_info, reverse, swh_object_icons, ) from swh.web.config import get_config _empty_snapshot_id = snapshot_identifier({"branches": {}}) def _get_branch(branches, branch_name, snapshot_id): """ Utility function to get a specific branch from a branches list. Its purpose is to get the default HEAD branch as some software origin (e.g those with svn type) does not have it. In that latter case, check if there is a master branch instead and returns it. """ filtered_branches = [b for b in branches if b["name"] == branch_name] if filtered_branches: return filtered_branches[0] elif branch_name == "HEAD": filtered_branches = [b for b in branches if b["name"].endswith("master")] if filtered_branches: return filtered_branches[0] elif branches: return branches[0] else: # case where a large branches list has been truncated snp = service.lookup_snapshot( snapshot_id, branches_from=branch_name, branches_count=1, target_types=["revision", "alias"], ) snp_branch, _ = process_snapshot_branches(snp) if snp_branch and snp_branch[0]["name"] == branch_name: branches.append(snp_branch[0]) return snp_branch[0] def _get_release(releases, release_name, snapshot_id): """ Utility function to get a specific release from a releases list. Returns None if the release can not be found in the list. """ filtered_releases = [r for r in releases if r["name"] == release_name] if filtered_releases: return filtered_releases[0] else: # case where a large branches list has been truncated try: # git origins have specific branches for releases snp = service.lookup_snapshot( snapshot_id, branches_from=f"refs/tags/{release_name}", branches_count=1, target_types=["release"], ) except NotFoundExc: snp = service.lookup_snapshot( snapshot_id, branches_from=release_name, branches_count=1, target_types=["release"], ) _, snp_release = process_snapshot_branches(snp) if snp_release and snp_release[0]["name"] == release_name: releases.append(snp_release[0]) return snp_release[0] def _branch_not_found( branch_type, branch, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id ): """ Utility function to raise an exception when a specified branch/release can not be found. """ if branch_type == "branch": branch_type = "Branch" branch_type_plural = "branches" target_type = "revision" else: branch_type = "Release" branch_type_plural = "releases" target_type = "release" if snapshot_id and snapshot_sizes[target_type] == 0: msg = "Snapshot with id %s has an empty list" " of %s!" % ( snapshot_id, branch_type_plural, ) elif snapshot_id: msg = "%s %s for snapshot with id %s" " not found!" % ( branch_type, branch, snapshot_id, ) elif visit_id and snapshot_sizes[target_type] == 0: msg = ( "Origin with url %s" " for visit with id %s has an empty list" " of %s!" % (origin_info["url"], visit_id, branch_type_plural) ) elif visit_id: msg = ( "%s %s associated to visit with" " id %s for origin with url %s" " not found!" % (branch_type, branch, visit_id, origin_info["url"]) ) elif snapshot_sizes[target_type] == 0: msg = ( "Origin with url %s" " for visit with timestamp %s has an empty list" " of %s!" % (origin_info["url"], timestamp, branch_type_plural) ) else: msg = ( "%s %s associated to visit with" " timestamp %s for origin with " "url %s not found!" % (branch_type, branch, timestamp, origin_info["url"]) ) raise NotFoundExc(escape(msg)) def process_snapshot_branches( snapshot: Dict[str, Any] ) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo]]: """ 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.service.lookup_snapshot` Returns: A tuple whose first member is the sorted list of branches targeting revisions and second member the sorted list of branches targeting releases """ 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, 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): releases[branch] = SnapshotReleaseInfo( name=release["name"], 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): branches[branch] = SnapshotBranchInfo( name=branch, revision=revision["id"], directory=revision["directory"], date=format_utc_iso_date(revision["date"]), message=revision["message"], url=None, ) releases_info = service.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 = service.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"] for branch_alias, branch_target in branch_aliases.items(): if branch_target in branches: branches[branch_alias] = copy(branches[branch_target]) else: snp = service.lookup_snapshot( snapshot["id"], branches_from=branch_target, branches_count=1 ) if snp and branch_target in snp["branches"]: if snp["branches"][branch_target] is None: continue target_type = snp["branches"][branch_target]["target_type"] target = snp["branches"][branch_target]["target"] if target_type == "revision": branches[branch_alias] = snp["branches"][branch_target] revision = service.lookup_revision(target) _add_branch_info(branch_alias, revision) elif target_type == "release": release = service.lookup_release(target) _add_release_info(branch_alias, release) 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 def get_snapshot_content( snapshot_id: str, ) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo]]: """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 two members. The first one is a list of dict describing the snapshot branches. The second one is a list of dict describing the snapshot releases. Raises: NotFoundExc if the snapshot does not exist """ cache_entry_id = "swh_snapshot_%s" % snapshot_id cache_entry = cache.get(cache_entry_id) if cache_entry: return cache_entry["branches"], cache_entry["releases"] branches: List[SnapshotBranchInfo] = [] releases: List[SnapshotReleaseInfo] = [] snapshot_content_max_size = get_config()["snapshot_content_max_size"] if snapshot_id: snapshot = service.lookup_snapshot( snapshot_id, branches_count=snapshot_content_max_size ) branches, releases = process_snapshot_branches(snapshot) cache.set(cache_entry_id, {"branches": branches, "releases": releases,}) return branches, releases 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]]: """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 two members. The first one is a list of dict describing the origin branches for the given visit. The second one is a list of dict describing the origin releases for the given visit. Raises: NotFoundExc if the origin or its visit are not found """ visit_info = get_origin_visit(origin_info, visit_ts, visit_id, snapshot_id) return get_snapshot_content(visit_info["snapshot"]) def 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 = service.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 = get_origin_visit_snapshot( origin_info, timestamp, visit_id, snapshot_id ) query_params["origin_url"] = origin_info["url"] origin_visits_url = reverse( "browse-origin-visits", query_params={"origin_url": origin_info["url"]} ) if timestamp is not None: query_params["timestamp"] = format_utc_iso_date( timestamp, "%Y-%m-%dT%H:%M:%SZ" ) visit_url = reverse("browse-origin-directory", query_params=query_params) visit_info["url"] = visit_url branches_url = reverse("browse-origin-branches", query_params=query_params) releases_url = reverse("browse-origin-releases", query_params=query_params) else: assert snapshot_id is not None branches, releases = get_snapshot_content(snapshot_id) url_args = {"snapshot_id": snapshot_id} branches_url = reverse("browse-snapshot-branches", url_args=url_args) releases_url = reverse("browse-snapshot-releases", url_args=url_args) releases = list(reversed(releases)) snapshot_sizes = service.lookup_snapshot_sizes(snapshot_id) is_empty = sum(snapshot_sizes.values()) == 0 swh_snp_id = swhid("snapshot", 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 = sum(snapshot_sizes.values()) if path is not None: query_params["path"] = path if snapshot_total_size and revision_id is not None: revision = service.lookup_revision(revision_id) root_directory = revision["directory"] branches.append( SnapshotBranchInfo( name=revision_id, revision=revision_id, directory=root_directory, date=revision["date"], message=revision["message"], url=None, ) ) branch_name = revision_id query_params["revision"] = revision_id elif snapshot_total_size and release_name: release = _get_release(releases, release_name, snapshot_id) if release is None: _branch_not_found( "release", release_name, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id, ) else: root_directory = release["directory"] revision_id = release["target"] release_id = release["id"] query_params["release"] = release_name elif snapshot_total_size: if branch_name: query_params["branch"] = branch_name branch = _get_branch(branches, branch_name or "HEAD", snapshot_id) if branch is None: _branch_not_found( "branch", branch_name, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id, ) else: branch_name = branch["name"] revision_id = branch["revision"] root_directory = branch["directory"] for b in branches: branch_query_params = dict(query_params) branch_query_params.pop("release", None) if b["name"] != b["revision"]: branch_query_params.pop("revision", None) branch_query_params["branch"] = b["name"] b["url"] = reverse( browse_view_name, url_args=url_args, query_params=branch_query_params ) for r in releases: release_query_params = dict(query_params) release_query_params.pop("branch", None) release_query_params.pop("revision", None) release_query_params["release"] = r["name"] r["url"] = reverse( browse_view_name, url_args=url_args, query_params=release_query_params, ) revision_info = None if revision_id: try: revision_info = service.lookup_revision(revision_id) except NotFoundExc: pass else: revision_info["date"] = format_utc_iso_date(revision_info["date"]) revision_info["committer_date"] = format_utc_iso_date( revision_info["committer_date"] ) if revision_info["message"]: message_lines = revision_info["message"].split("\n") revision_info["message_header"] = message_lines[0] else: revision_info["message_header"] = "" snapshot_context = SnapshotContext( branch=branch_name, branches=branches, branches_url=branches_url, is_empty=is_empty, origin_info=origin_info, origin_visits_url=origin_visits_url, release=release_name, release_id=release_id, query_params=query_params, releases=releases, releases_url=releases_url, revision_id=revision_id, revision_info=revision_info, root_directory=root_directory, snapshot_id=snapshot_id, snapshot_sizes=snapshot_sizes, snapshot_swhid=swh_snp_id, url_args=url_args, visit_info=visit_info, ) if revision_info: revision_info["revision_url"] = gen_revision_url(revision_id, snapshot_context) return snapshot_context def _build_breadcrumbs(snapshot_context: SnapshotContext, path: str): origin_info = snapshot_context["origin_info"] url_args = snapshot_context["url_args"] query_params = dict(snapshot_context["query_params"]) root_directory = snapshot_context["root_directory"] path_info = gen_path_info(path) if origin_info: browse_view_name = "browse-origin-directory" else: browse_view_name = "browse-snapshot-directory" breadcrumbs = [] if root_directory: query_params.pop("path", None) breadcrumbs.append( { "name": root_directory[:7], "url": reverse( browse_view_name, url_args=url_args, query_params=query_params ), } ) for pi in path_info: query_params["path"] = pi["path"] breadcrumbs.append( { "name": pi["name"], "url": reverse( browse_view_name, url_args=url_args, query_params=query_params ), } ) return breadcrumbs def _check_origin_url(snapshot_id, origin_url): if snapshot_id is None and origin_url is None: raise BadInputExc("An origin URL must be provided as query parameter.") def browse_snapshot_directory( request, snapshot_id=None, origin_url=None, timestamp=None, path=None ): """ Django view implementation for browsing a directory in a snapshot context. """ try: _check_origin_url(snapshot_id, origin_url) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), path=path, browse_context="directory", branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), ) root_directory = snapshot_context["root_directory"] sha1_git = root_directory if root_directory and path: dir_info = service.lookup_directory_with_path(root_directory, path) sha1_git = dir_info["target"] dirs = [] files = [] if sha1_git: dirs, files = get_directory_entries(sha1_git) except Exception as exc: return handle_view_exception(request, exc) 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"] f["length"] = filesizeformat(f["length"]) if f["name"].lower().startswith("readme"): readmes[f["name"]] = f["checksums"]["sha1"] readme_name, readme_url, readme_html = get_readme_to_display(readmes) 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) sum_file_sizes = filesizeformat(sum_file_sizes) dir_path = "/" + path browse_dir_link = gen_directory_link(sha1_git) browse_rev_link = gen_revision_link(revision_id) browse_snp_link = gen_snapshot_link(snapshot_id) revision_found = True if sha1_git is None and revision_id is not None: try: service.lookup_revision(revision_id) except NotFoundExc: revision_found = False swh_objects = [ SWHObjectInfo(object_type=DIRECTORY, object_id=sha1_git), SWHObjectInfo(object_type=REVISION, object_id=revision_id), SWHObjectInfo(object_type=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"] browse_rel_link = None if release_id: swh_objects.append(SWHObjectInfo(object_type=RELEASE, object_id=release_id)) browse_rel_link = gen_release_link(release_id) dir_metadata = DirectoryMetadata( object_type=DIRECTORY, object_id=sha1_git, directory=sha1_git, directory_url=browse_dir_link, 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, revision_url=browse_rev_link, release=release_id, release_url=browse_rel_link, snapshot=snapshot_id, snapshot_url=browse_snp_link, origin_url=origin_url, visit_date=visit_date, visit_type=visit_type, ) vault_cooking = { "directory_context": True, "directory_id": sha1_git, "revision_context": True, "revision_id": revision_id, } 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"]: 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, }, ) def browse_snapshot_content( request, snapshot_id=None, origin_url=None, timestamp=None, path=None, selected_language=None, ): """ Django view implementation for browsing a content in a snapshot context. """ try: _check_origin_url(snapshot_id, origin_url) if path is None: raise BadInputExc("The path of a content must be given as query parameter.") snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), path=path, browse_context="content", branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), ) root_directory = snapshot_context["root_directory"] sha1_git = None query_string = None content_data = {} directory_id = None split_path = path.split("/") filename = split_path[-1] filepath = path[: -len(filename)] if root_directory: content_info = service.lookup_directory_with_path(root_directory, path) sha1_git = content_info["target"] query_string = "sha1_git:" + sha1_git content_data = request_content(query_string, raise_if_unavailable=False) if filepath: dir_info = service.lookup_directory_with_path(root_directory, filepath) directory_id = dir_info["target"] else: directory_id = root_directory except Exception as exc: return handle_view_exception(request, exc) revision_id = snapshot_context["revision_id"] origin_info = snapshot_context["origin_info"] visit_info = snapshot_context["visit_info"] snapshot_id = snapshot_context["snapshot_id"] if content_data.get("raw_data") is not None: content_display_data = prepare_content_for_display( content_data["raw_data"], content_data["mimetype"], path ) content_data.update(content_display_data) # Override language with user-selected language if selected_language is not None: content_data["language"] = selected_language available_languages = None if content_data.get("mimetype") is not None and "text/" in content_data["mimetype"]: available_languages = highlightjs.get_supported_languages() breadcrumbs = _build_breadcrumbs(snapshot_context, filepath) breadcrumbs.append({"name": filename, "url": None}) browse_content_link = gen_content_link(sha1_git) content_raw_url = None if query_string: content_raw_url = reverse( "browse-content-raw", url_args={"query_string": query_string}, query_params={"filename": filename}, ) browse_rev_link = gen_revision_link(revision_id) browse_dir_link = gen_directory_link(directory_id) content_checksums = content_data.get("checksums", {}) swh_objects = [ SWHObjectInfo(object_type=CONTENT, object_id=content_checksums.get("sha1_git")), SWHObjectInfo(object_type=DIRECTORY, object_id=directory_id), SWHObjectInfo(object_type=REVISION, object_id=revision_id), SWHObjectInfo(object_type=SNAPSHOT, object_id=snapshot_id), ] 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"] browse_rel_link = None if release_id: swh_objects.append(SWHObjectInfo(object_type=RELEASE, object_id=release_id)) browse_rel_link = gen_release_link(release_id) content_metadata = ContentMetadata( object_type=CONTENT, object_id=content_checksums.get("sha1_git"), sha1=content_checksums.get("sha1"), sha1_git=content_checksums.get("sha1_git"), sha256=content_checksums.get("sha256"), blake2s256=content_checksums.get("blake2s256"), content_url=browse_content_link, mimetype=content_data.get("mimetype"), encoding=content_data.get("encoding"), size=filesizeformat(content_data.get("length", 0)), language=content_data.get("language"), - licenses=content_data.get("licenses"), root_directory=root_directory, path=f"/{filepath}", filename=filename, directory=directory_id, directory_url=browse_dir_link, revision=revision_id, revision_url=browse_rev_link, release=release_id, release_url=browse_rel_link, snapshot=snapshot_id, snapshot_url=gen_snapshot_link(snapshot_id), origin_url=origin_url, visit_date=visit_date, visit_type=visit_type, ) swhids_info = get_swhids_info(swh_objects, snapshot_context, content_metadata) content_path = "/".join([bc["name"] for bc in breadcrumbs]) context_found = "snapshot: %s" % snapshot_context["snapshot_id"] if origin_info: context_found = "origin: %s" % origin_info["url"] heading = "Content - %s - %s - %s" % ( content_path, snapshot_context["branch"], context_found, ) top_right_link = None if not snapshot_context["is_empty"]: top_right_link = { "url": content_raw_url, "icon": swh_object_icons["content"], "text": "Raw File", } return render( request, "browse/content.html", { "heading": heading, "swh_object_name": "Content", "swh_object_metadata": content_metadata, "content": content_data.get("content_data"), "content_size": content_data.get("length"), "max_content_size": content_display_max_size, "filename": filename, "encoding": content_data.get("encoding"), "mimetype": content_data.get("mimetype"), "language": content_data.get("language"), "available_languages": available_languages, "breadcrumbs": breadcrumbs if root_directory else [], "top_right_link": top_right_link, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": True, "swhids_info": swhids_info, "error_code": content_data.get("error_code"), "error_message": content_data.get("error_message"), "error_description": content_data.get("error_description"), }, status=content_data.get("error_code", 200), ) PER_PAGE = 100 def browse_snapshot_log(request, snapshot_id=None, origin_url=None, timestamp=None): """ Django view implementation for browsing a revision history in a snapshot context. """ try: _check_origin_url(snapshot_id, origin_url) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), browse_context="log", branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), ) revision_id = snapshot_context["revision_id"] per_page = int(request.GET.get("per_page", PER_PAGE)) offset = int(request.GET.get("offset", 0)) revs_ordering = request.GET.get("revs_ordering", "committer_date") session_key = "rev_%s_log_ordering_%s" % (revision_id, revs_ordering) rev_log_session = request.session.get(session_key, None) rev_log = [] revs_walker_state = None if rev_log_session: rev_log = rev_log_session["rev_log"] revs_walker_state = rev_log_session["revs_walker_state"] if len(rev_log) < offset + per_page: revs_walker = service.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 = service.lookup_revision_multiple(revs) request.session[session_key] = { "rev_log": rev_log, "revs_walker_state": revs_walker_state, } except Exception as exc: return handle_view_exception(request, exc) origin_info = snapshot_context["origin_info"] visit_info = snapshot_context["visit_info"] url_args = snapshot_context["url_args"] query_params = snapshot_context["query_params"] snapshot_id = snapshot_context["snapshot_id"] query_params["per_page"] = per_page revs_ordering = request.GET.get("revs_ordering", "") query_params["revs_ordering"] = revs_ordering or None if origin_info: browse_view_name = "browse-origin-log" else: browse_view_name = "browse-snapshot-log" prev_log_url = None if len(rev_log) > offset + per_page: query_params["offset"] = offset + per_page prev_log_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) next_log_url = None if offset != 0: query_params["offset"] = offset - per_page next_log_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) revision_log_data = format_log_entries(revision_log, per_page, snapshot_context) browse_rev_link = gen_revision_link(revision_id) browse_log_link = gen_revision_log_link(revision_id) browse_snp_link = gen_snapshot_link(snapshot_id) revision_metadata = { "context-independent revision": browse_rev_link, "context-independent revision history": browse_log_link, "context-independent snapshot": browse_snp_link, "snapshot": snapshot_id, } if origin_info: revision_metadata["origin url"] = origin_info["url"] revision_metadata["origin visit date"] = format_utc_iso_date(visit_info["date"]) revision_metadata["origin visit type"] = visit_info["type"] swh_objects = [ SWHObjectInfo(object_type=REVISION, object_id=revision_id), SWHObjectInfo(object_type=SNAPSHOT, object_id=snapshot_id), ] release_id = snapshot_context["release_id"] if release_id: swh_objects.append(SWHObjectInfo(object_type=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 ): """ Django view implementation for browsing a list of branches in a snapshot context. """ try: _check_origin_url(snapshot_id, origin_url) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), ) branches_bc = request.GET.get("branches_breadcrumbs", "") branches_bc = branches_bc.split(",") if branches_bc else [] branches_from = branches_bc[-1] if branches_bc else "" origin_info = snapshot_context["origin_info"] url_args = snapshot_context["url_args"] query_params = snapshot_context["query_params"] if origin_info: browse_view_name = "browse-origin-directory" else: browse_view_name = "browse-snapshot-directory" snapshot = service.lookup_snapshot( snapshot_context["snapshot_id"], branches_from, PER_PAGE + 1, target_types=["revision", "alias"], ) displayed_branches, _ = process_snapshot_branches(snapshot) except Exception as exc: return handle_view_exception(request, exc) 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["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, }, ) def browse_snapshot_releases( request, snapshot_id=None, origin_url=None, timestamp=None ): """ Django view implementation for browsing a list of releases in a snapshot context. """ try: _check_origin_url(snapshot_id, origin_url) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), ) rel_bc = request.GET.get("releases_breadcrumbs", "") rel_bc = rel_bc.split(",") if rel_bc else [] rel_from = rel_bc[-1] if rel_bc else "" origin_info = snapshot_context["origin_info"] url_args = snapshot_context["url_args"] query_params = snapshot_context["query_params"] snapshot = service.lookup_snapshot( snapshot_context["snapshot_id"], rel_from, PER_PAGE + 1, target_types=["release", "alias"], ) _, displayed_releases = process_snapshot_branches(snapshot) except Exception as exc: return handle_view_exception(request, exc) for release in displayed_releases: query_params_tgt = {"snapshot": snapshot_id} if origin_info: query_params_tgt["origin_url"] = origin_info["url"] release_url = reverse( "browse-release", url_args={"sha1_git": release["id"]}, query_params=query_params_tgt, ) target_url = "" 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, ) release["release_url"] = release_url release["target_url"] = target_url 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["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, }, ) diff --git a/swh/web/browse/utils.py b/swh/web/browse/utils.py index 8d1f80d8..ac0dbfea 100644 --- a/swh/web/browse/utils.py +++ b/swh/web/browse/utils.py @@ -1,742 +1,736 @@ # Copyright (C) 2017-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import base64 import stat import textwrap from threading import Lock import magic import sentry_sdk from django.core.cache import cache from django.utils.html import escape from django.utils.safestring import mark_safe from swh.web.common import highlightjs, service from swh.web.common.exc import http_status_code_message from swh.web.common.utils import ( browsers_supported_image_mimes, format_utc_iso_date, reverse, rst_to_html, ) from swh.web.config import get_config def get_directory_entries(sha1_git): """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 """ cache_entry_id = "directory_entries_%s" % sha1_git cache_entry = cache.get(cache_entry_id) if cache_entry: return cache_entry entries = list(service.lookup_directory(sha1_git)) 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"]) cache.set(cache_entry_id, (dirs, files)) return dirs, files _lock = Lock() def get_mimetype_and_encoding_for_content(content): """Function that returns the mime type and the encoding associated to a content buffer using the magic module under the hood. Args: content (bytes): a content buffer Returns: A tuple (mimetype, encoding), for instance ('text/plain', 'us-ascii'), associated to the provided content. """ 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, encoding, content_data): # encode textual content to utf-8 if needed if mimetype.startswith("text/"): # probably a malformed UTF-8 content, re-encode it # by replacing invalid chars with a substitution one if encoding == "unknown-8bit": 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 as exc: sentry_sdk.capture_exception(exc) 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, raise_if_unavailable=True, re_encode=True, ): """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 Raises: NotFoundExc if the content is not found """ content_data = service.lookup_content(query_string) filetype = None language = None - license = None # requests to the indexer db may fail so properly handle # those cases in order to avoid content display errors try: filetype = service.lookup_content_filetype(query_string) language = service.lookup_content_language(query_string) - license = service.lookup_content_license(query_string) except Exception as exc: sentry_sdk.capture_exception(exc) mimetype = "unknown" encoding = "unknown" if filetype: mimetype = filetype["mimetype"] encoding = filetype["encoding"] # workaround when encountering corrupted data due to implicit # conversion from bytea to text in the indexer db (see T818) # TODO: Remove that code when all data have been correctly converted if mimetype.startswith("\\"): filetype = None content_data["error_code"] = 200 content_data["error_message"] = "" content_data["error_description"] = "" if not max_size or content_data["length"] < max_size: try: content_raw = service.lookup_content_raw(query_string) except Exception as exc: if raise_if_unavailable: raise exc else: sentry_sdk.capture_exception(exc) content_data["raw_data"] = None content_data["error_code"] = 404 content_data["error_description"] = ( "The bytes of the content are currently not available " "in the archive." ) content_data["error_message"] = http_status_code_message[ content_data["error_code"] ] else: content_data["raw_data"] = content_raw["data"] if not filetype: mimetype, encoding = get_mimetype_and_encoding_for_content( content_data["raw_data"] ) 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" - if license: - content_data["licenses"] = ", ".join(license["facts"][0]["licenses"]) - else: - content_data["licenses"] = "not detected" return content_data def prepare_content_for_display(content_data, mime_type, path): """Function that prepares a content for HTML display. The function tries to associate a programming language to a content in order to perform syntax highlighting client-side using highlightjs. The language is determined using either the content filename or its mime type. If the mime type corresponds to an image format supported by web browsers, the content will be encoded in base64 for displaying the image. Args: content_data (bytes): raw bytes of the content mime_type (string): mime type of the content path (string): path of the content including filename Returns: A dict containing the content bytes (possibly different from the one provided as parameter if it is an image) under the key 'content_data and the corresponding highlightjs language class under the key 'language'. """ language = highlightjs.get_hljs_language_from_filename(path) if not language: language = highlightjs.get_hljs_language_from_mime_type(mime_type) if not language: language = "nohighlight" elif mime_type.startswith("application/"): mime_type = mime_type.replace("application/", "text/") if mime_type.startswith("image/"): if mime_type in browsers_supported_image_mimes: content_data = base64.b64encode(content_data).decode("ascii") if mime_type.startswith("image/svg"): mime_type = "image/svg+xml" if mime_type.startswith("text/"): content_data = content_data.decode("utf-8", errors="replace") return {"content_data": content_data, "language": language, "mimetype": mime_type} def gen_link(url, link_text=None, link_attrs=None): """ 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, if not provided the url will be used link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ attrs = " " 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 = {} 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"] if "visit_id" in snp_query_params: query_params["visit_id"] = 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"] 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): """ Utility function for generating an url to a revision. Args: revision_id (str): a revision id snapshot_context (dict): if provided, generate snapshot-dependent browsing url Returns: str: The url to browse the revision """ query_params = _snapshot_context_query_params(snapshot_context) query_params.pop("revision", 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"}, ): """ 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 characters for the link text snapshot_context (dict): if provided, generate snapshot-dependent browsing link link_text (str): optional text for the generated link (the revision id will be used by default) link_attrs (dict): 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"}, ): """ 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 (the directory id will be used by default) link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ 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"}, ): """ 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 (the snapshot id will be used by default) link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ 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"}, ): """ 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 (the content sha1_git will be used by default) link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ 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): """ 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 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"}, ): """ 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 browsing link link_text (str): optional text to use for the generated link (the revision id will be used by default) link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ 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): """ Utility function for generating a mail link to a person to insert in Django templates. Args: person (dict): dictionary containing person data (*name*, *email*, *fullname*) link_text (str): 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"}, ): """ 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 (the release id will be used by default) link_attrs (dict): optional attributes (e.g. class) to add to the link Returns: An HTML link in the form 'link_text' """ 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): """ 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 browsing link """ revision_log_data = [] for i, rev in enumerate(revision_log): 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): """ Process a list of readme files found in a directory in order to find the adequate one to display. Args: readmes: a list of dict where keys are readme file names and values are readme sha1s Returns: A tuple (readme_name, readme_sha1) """ readme_name = None readme_url = None readme_sha1 = None readme_html = None lc_readmes = {k.lower(): {"orig_name": k, "sha1": v} for k, v in readmes.items()} # look for readme names according to the preference order # defined by the _common_readme_names list for common_readme_name in _common_readme_names: if common_readme_name in lc_readmes: readme_name = lc_readmes[common_readme_name]["orig_name"] readme_sha1 = lc_readmes[common_readme_name]["sha1"] readme_url = reverse( "browse-content-raw", 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"): cache_entry_id = "readme_%s" % readme_sha1 cache_entry = cache.get(cache_entry_id) if cache_entry: readme_html = cache_entry else: try: rst_doc = request_content(readme_sha1) readme_html = rst_to_html(rst_doc["raw_data"]) cache.set(cache_entry_id, readme_html) except Exception as exc: sentry_sdk.capture_exception(exc) readme_html = "Readme bytes are not available" return readme_name, readme_url, readme_html diff --git a/swh/web/browse/views/content.py b/swh/web/browse/views/content.py index 1463a8b7..9d17693c 100644 --- a/swh/web/browse/views/content.py +++ b/swh/web/browse/views/content.py @@ -1,408 +1,407 @@ # Copyright (C) 2017-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import difflib from distutils.util import strtobool import sentry_sdk from django.http import HttpResponse, JsonResponse from django.shortcuts import render from django.template.defaultfilters import filesizeformat from swh.model.hashutil import hash_to_hex from swh.model.identifiers import CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import ( content_display_max_size, gen_directory_link, gen_link, prepare_content_for_display, request_content, ) from swh.web.common import highlightjs, query, service from swh.web.common.exc import NotFoundExc, handle_view_exception from swh.web.common.identifiers import get_swhids_info from swh.web.common.typing import ContentMetadata, SWHObjectInfo from swh.web.common.utils import gen_path_info, reverse, swh_object_icons @browse_route( r"content/(?P[0-9a-z_:]*[0-9a-f]+.)/raw/", view_name="browse-content-raw", checksum_args=["query_string"], ) def content_raw(request, query_string): """Django view that produces a raw display of a content identified by its hash value. The url that points to it is :http:get:`/browse/content/[(algo_hash):](hash)/raw/` """ try: 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) except Exception as exc: return handle_view_exception(request, exc) filename = request.GET.get("filename", None) if not filename: filename = "%s_%s" % (algo, checksum) if ( content_data["mimetype"].startswith("text/") or content_data["mimetype"] == "inode/x-empty" ): response = HttpResponse(content_data["raw_data"], content_type="text/plain") response["Content-disposition"] = "filename=%s" % filename else: response = HttpResponse( content_data["raw_data"], content_type="application/octet-stream" ) response["Content-disposition"] = "attachment; filename=%s" % filename return response _auto_diff_size_limit = 20000 @browse_route( r"content/(?P.*)/diff/(?P.*)", view_name="diff-contents", ) def _contents_diff(request, from_query_string, to_query_string): """ Browse endpoint used to compute unified diffs between two contents. Diffs are generated only if the two contents are textual. By default, diffs whose size are greater than 20 kB will not be generated. To force the generation of large diffs, the 'force' boolean query parameter must be used. Args: request: input django http request from_query_string: a string of the form "[ALGO_HASH:]HASH" where optional ALGO_HASH can be either ``sha1``, ``sha1_git``, ``sha256``, or ``blake2s256`` (default to ``sha1``) and HASH the hexadecimal representation of the hash value identifying the first content to_query_string: same as above for identifying the second content Returns: A JSON object containing the unified diff. """ diff_data = {} content_from = None content_to = None content_from_size = 0 content_to_size = 0 content_from_lines = [] content_to_lines = [] force = request.GET.get("force", "false") path = request.GET.get("path", None) language = "nohighlight" force = bool(strtobool(force)) if from_query_string == to_query_string: diff_str = "File renamed without changes" else: try: text_diff = True if from_query_string: content_from = request_content(from_query_string, max_size=None) content_from_display_data = prepare_content_for_display( content_from["raw_data"], content_from["mimetype"], path ) language = content_from_display_data["language"] content_from_size = content_from["length"] if not ( content_from["mimetype"].startswith("text/") or content_from["mimetype"] == "inode/x-empty" ): text_diff = False if text_diff and to_query_string: content_to = request_content(to_query_string, max_size=None) content_to_display_data = prepare_content_for_display( content_to["raw_data"], content_to["mimetype"], path ) language = content_to_display_data["language"] content_to_size = content_to["length"] if not ( content_to["mimetype"].startswith("text/") or content_to["mimetype"] == "inode/x-empty" ): text_diff = False diff_size = abs(content_to_size - content_from_size) if not text_diff: diff_str = "Diffs are not generated for non textual content" language = "nohighlight" elif not force and diff_size > _auto_diff_size_limit: diff_str = "Large diffs are not automatically computed" language = "nohighlight" else: if content_from: content_from_lines = ( content_from["raw_data"].decode("utf-8").splitlines(True) ) if content_from_lines and content_from_lines[-1][-1] != "\n": content_from_lines[-1] += "[swh-no-nl-marker]\n" if content_to: content_to_lines = ( content_to["raw_data"].decode("utf-8").splitlines(True) ) if content_to_lines and content_to_lines[-1][-1] != "\n": content_to_lines[-1] += "[swh-no-nl-marker]\n" diff_lines = difflib.unified_diff(content_from_lines, content_to_lines) diff_str = "".join(list(diff_lines)[2:]) except Exception as exc: sentry_sdk.capture_exception(exc) diff_str = str(exc) diff_data["diff_str"] = diff_str diff_data["language"] = language return JsonResponse(diff_data) @browse_route( r"content/(?P[0-9a-z_:]*[0-9a-f]+.)/", view_name="browse-content", checksum_args=["query_string"], ) def content_display(request, query_string): """Django view that produces an HTML display of a content identified by its hash value. The url that points to it is :http:get:`/browse/content/[(algo_hash):](hash)/` """ try: algo, checksum = query.parse_hash(query_string) checksum = hash_to_hex(checksum) content_data = request_content(query_string, raise_if_unavailable=False) origin_url = request.GET.get("origin_url") selected_language = request.GET.get("language") if not origin_url: origin_url = request.GET.get("origin") snapshot_id = request.GET.get("snapshot") path = request.GET.get("path") snapshot_context = None if origin_url is not None or snapshot_id is not None: try: snapshot_context = get_snapshot_context( origin_url=origin_url, snapshot_id=snapshot_id, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), path=path, browse_context=CONTENT, ) except NotFoundExc as e: if str(e).startswith("Origin"): raw_cnt_url = reverse( "browse-content", url_args={"query_string": query_string} ) error_message = ( "The Software Heritage archive has a content " "with the hash you provided but the origin " "mentioned in your request appears broken: %s. " "Please check the URL and try again.\n\n" "Nevertheless, you can still browse the content " "without origin information: %s" % (gen_link(origin_url), gen_link(raw_cnt_url)) ) raise NotFoundExc(error_message) else: raise e except Exception as exc: return handle_view_exception(request, exc) content = None language = None mimetype = None if content_data["raw_data"] is not None: content_display_data = prepare_content_for_display( content_data["raw_data"], content_data["mimetype"], path ) content = content_display_data["content_data"] language = content_display_data["language"] mimetype = content_display_data["mimetype"] # 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 directory_url = None root_dir = None if snapshot_context: root_dir = snapshot_context.get("root_directory") query_params = snapshot_context["query_params"] if snapshot_context else {} breadcrumbs = [] if path: split_path = path.split("/") root_dir = root_dir or split_path[0] filename = split_path[-1] if root_dir != path: path = path.replace(root_dir + "/", "") path = path[: -len(filename)] path_info = gen_path_info(path) query_params.pop("path", None) dir_url = reverse( "browse-directory", url_args={"sha1_git": root_dir}, query_params=query_params, ) breadcrumbs.append({"name": root_dir[:7], "url": dir_url}) for pi in path_info: query_params["path"] = pi["path"] dir_url = reverse( "browse-directory", url_args={"sha1_git": root_dir}, query_params=query_params, ) breadcrumbs.append({"name": pi["name"], "url": dir_url}) breadcrumbs.append({"name": filename, "url": None}) if path and root_dir != path: try: dir_info = service.lookup_directory_with_path(root_dir, path) directory_id = dir_info["target"] except Exception as exc: return handle_view_exception(request, exc) elif root_dir != path: directory_id = root_dir else: root_dir = None if directory_id: directory_url = gen_directory_link(directory_id) query_params = {"filename": filename} content_checksums = content_data["checksums"] content_url = reverse( "browse-content", url_args={"query_string": f'sha1_git:{content_checksums["sha1_git"]}'}, ) content_raw_url = reverse( "browse-content-raw", url_args={"query_string": query_string}, query_params=query_params, ) content_metadata = ContentMetadata( object_type=CONTENT, object_id=content_checksums["sha1_git"], sha1=content_checksums["sha1"], sha1_git=content_checksums["sha1_git"], sha256=content_checksums["sha256"], blake2s256=content_checksums["blake2s256"], content_url=content_url, mimetype=content_data["mimetype"], encoding=content_data["encoding"], size=filesizeformat(content_data["length"]), language=content_data["language"], - licenses=content_data["licenses"], root_directory=root_dir, path=f"/{path}" if path else None, filename=filename or "", directory=directory_id, directory_url=directory_url, revision=None, release=None, snapshot=None, origin_url=origin_url, ) swh_objects = [ SWHObjectInfo(object_type=CONTENT, object_id=content_checksums["sha1_git"]) ] if directory_id: swh_objects.append(SWHObjectInfo(object_type=DIRECTORY, object_id=directory_id)) if snapshot_context: swh_objects.append( SWHObjectInfo( object_type=REVISION, object_id=snapshot_context["revision_id"] ) ) swh_objects.append( SWHObjectInfo( object_type=SNAPSHOT, object_id=snapshot_context["snapshot_id"] ) ) if snapshot_context["release_id"]: swh_objects.append( SWHObjectInfo( object_type=RELEASE, object_id=snapshot_context["release_id"] ) ) swhids_info = get_swhids_info( swh_objects, snapshot_context, extra_context=content_metadata, ) heading = "Content - %s" % content_checksums["sha1_git"] if breadcrumbs: content_path = "/".join([bc["name"] for bc in breadcrumbs]) heading += " - %s" % content_path return render( request, "browse/content.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"], "swh_object_name": "Content", "swh_object_metadata": content_metadata, "content": content, "content_size": content_data["length"], "max_content_size": content_display_max_size, "filename": filename, "encoding": content_data["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": content_data["error_code"], "error_message": content_data["error_message"], "error_description": content_data["error_description"], }, status=content_data["error_code"], ) diff --git a/swh/web/common/typing.py b/swh/web/common/typing.py index 6a027ac2..a0c8b4dd 100644 --- a/swh/web/common/typing.py +++ b/swh/web/common/typing.py @@ -1,224 +1,223 @@ # Copyright (C) 2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from 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 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: Optional[str] """branch name""" revision: str """branch heading revision""" url: Optional[str] """optional browse URL (content, directory, ...) scoped to branch""" class SnapshotReleaseInfo(TypedDict): branch_name: str """branch name associated to release in snapshot""" date: str """release date""" directory: Optional[str] """optional directory associatd to the release""" id: str """release identifier""" message: str """release message""" name: str """release name""" 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""" 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_id: Optional[str] """optional release identifier set when browsing snapshot in that scope""" releases: List[SnapshotReleaseInfo] """list of snapshot releases (possibly truncated)""" releases_url: str """snapshot releases list browse URL""" revision_id: Optional[str] """optional revision identifier set when browsing snapshot in that scope""" revision_info: Optional[Dict[str, Any]] """optional revision info set when browsing snapshot in that scope""" root_directory: Optional[str] """optional root directory identifier set when browsing snapshot content""" snapshot_id: str """snapshot identifier""" snapshot_sizes: Dict[str, int] """snapshot sizes grouped by branch target type""" snapshot_swhid: str """snapshot SWHID""" url_args: Dict[str, Any] """common URL arguments when browsing snapshot content""" visit_info: Optional[OriginVisitInfo] """optional origin visit info associated to the snapshot""" class SWHObjectInfo(TypedDict): object_type: str object_id: str class SWHIDContext(TypedDict, total=False): origin: str anchor: str visit: str path: str lines: str class SWHIDInfo(SWHObjectInfo): swhid: str swhid_url: str context: SWHIDContext swhid_with_context: Optional[str] swhid_with_context_url: Optional[str] class SWHObjectInfoMetadata(TypedDict, total=False): origin_url: Optional[str] visit_date: Optional[str] visit_type: Optional[str] directory_url: Optional[str] revision_url: Optional[str] release_url: Optional[str] snapshot_url: Optional[str] class ContentMetadata(SWHObjectInfo, SWHObjectInfoMetadata): sha1: str sha1_git: str sha256: str blake2s256: str content_url: str mimetype: str encoding: str size: str language: str - licenses: 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: str root_directory: Optional[str] path: str revision: Optional[str] revision_found: Optional[bool] release: Optional[str] snapshot: Optional[str] class ReleaseMetadata(SWHObjectInfo, SWHObjectInfoMetadata): release: str author: str author_url: str date: str name: str synthetic: bool target: str target_type: str target_url: 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]