diff --git a/swh/web/api/views/directory.py b/swh/web/api/views/directory.py index 00d3c95d..1a8750e9 100644 --- a/swh/web/api/views/directory.py +++ b/swh/web/api/views/directory.py @@ -1,93 +1,93 @@ # Copyright (C) 2015-2019 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from swh.web.api import utils from swh.web.api.apidoc import api_doc, format_docstring from swh.web.api.apiurls import api_route from swh.web.api.views.utils import api_lookup from swh.web.common import archive @api_route( r"/directory/(?P[0-9a-f]+)/", "api-1-directory", checksum_args=["sha1_git"], ) @api_route( r"/directory/(?P[0-9a-f]+)/(?P.+)/", "api-1-directory", checksum_args=["sha1_git"], ) @api_doc("/directory/") @format_docstring() def api_directory(request, sha1_git, path=None): """ .. http:get:: /api/1/directory/(sha1_git)/[(path)/] Get information about directory objects. Directories are identified by **sha1** checksums, compatible with Git directory identifiers. - See :func:`swh.model.identifiers.directory_identifier` in our data model + See :func:`swh.model.git_objects.directory_git_object` in our data model module for details about how they are computed. When given only a directory identifier, this endpoint returns information about the directory itself, returning its content (usually a list of directory entries). When given a directory identifier and a path, this endpoint returns information about the directory entry pointed by the relative path, starting path resolution from the given directory. :param string sha1_git: hexadecimal representation of the directory **sha1_git** identifier :param string path: optional parameter to get information about the directory entry pointed by that relative path {common_headers} :>jsonarr object checksums: object holding the computed checksum values for a directory entry (only for file entries) :>jsonarr string dir_id: **sha1_git** identifier of the requested directory :>jsonarr number length: length of a directory entry in bytes (only for file entries) for getting information about the content MIME type :>jsonarr string name: the directory entry name :>jsonarr number perms: permissions for the directory entry :>jsonarr string target: **sha1_git** identifier of the directory entry :>jsonarr string target_url: link to :http:get:`/api/1/content/[(hash_type):](hash)/` or :http:get:`/api/1/directory/(sha1_git)/[(path)/]` depending on the directory entry type :>jsonarr string type: the type of the directory entry, can be either ``dir``, ``file`` or ``rev`` :statuscode 200: no error :statuscode 400: an invalid **hash_type** or **hash** has been provided :statuscode 404: requested directory can not be found in the archive **Example:** .. parsed-literal:: :swh_web_api:`directory/977fc4b98c0e85816348cebd3b12026407c368b6/` """ if path: error_msg_path = ( "Entry with path %s relative to directory " "with sha1_git %s not found." ) % (path, sha1_git) return api_lookup( archive.lookup_directory_with_path, sha1_git, path, notfound_msg=error_msg_path, enrich_fn=utils.enrich_directory_entry, request=request, ) else: error_msg_nopath = "Directory with sha1_git %s not found." % sha1_git return api_lookup( archive.lookup_directory, sha1_git, notfound_msg=error_msg_nopath, enrich_fn=utils.enrich_directory_entry, request=request, ) diff --git a/swh/web/api/views/graph.py b/swh/web/api/views/graph.py index c18a8216..89846537 100644 --- a/swh/web/api/views/graph.py +++ b/swh/web/api/views/graph.py @@ -1,160 +1,160 @@ # 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 distutils.util import strtobool import json from typing import Dict, Iterator, Union import requests from django.http.response import StreamingHttpResponse from rest_framework.decorators import renderer_classes from rest_framework.request import Request from rest_framework.response import Response from swh.model.hashutil import hash_to_hex -from swh.model.identifiers import ExtendedObjectType, ExtendedSWHID from swh.model.model import Sha1Git +from swh.model.swhids import ExtendedObjectType, ExtendedSWHID from swh.web.api.apidoc import api_doc from swh.web.api.apiurls import api_route from swh.web.api.renderers import PlainTextRenderer from swh.web.common import archive from swh.web.config import SWH_WEB_INTERNAL_SERVER_NAME, get_config API_GRAPH_PERM = "swh.web.api.graph" def _resolve_origin_swhid(swhid: str, origin_urls: Dict[Sha1Git, str]) -> str: """ Resolve origin url from its swhid sha1 representation. """ parsed_swhid = ExtendedSWHID.from_string(swhid) if parsed_swhid.object_type == ExtendedObjectType.ORIGIN: if parsed_swhid.object_id in origin_urls: return origin_urls[parsed_swhid.object_id] else: origin_info = list( archive.lookup_origins_by_sha1s([hash_to_hex(parsed_swhid.object_id)]) )[0] assert origin_info is not None origin_urls[parsed_swhid.object_id] = origin_info["url"] return origin_info["url"] else: return swhid def _resolve_origin_swhids_in_graph_response( response: requests.Response, ) -> Iterator[bytes]: """ Resolve origin urls from their swhid sha1 representations in graph service responses. """ content_type = response.headers["Content-Type"] origin_urls: Dict[Sha1Git, str] = {} if content_type == "application/x-ndjson": for line in response.iter_lines(): swhids = json.loads(line.decode("utf-8")) processed_line = [] for swhid in swhids: processed_line.append(_resolve_origin_swhid(swhid, origin_urls)) yield (json.dumps(processed_line) + "\n").encode() elif content_type == "text/plain": for line in response.iter_lines(): processed_line = [] swhids = line.decode("utf-8").split(" ") for swhid in swhids: processed_line.append(_resolve_origin_swhid(swhid, origin_urls)) yield (" ".join(processed_line) + "\n").encode() else: for line in response.iter_lines(): yield line + b"\n" @api_route(r"/graph/", "api-1-graph-doc") @api_doc("/graph/") def api_graph(request: Request) -> None: """ .. http:get:: /api/1/graph/(graph_query)/ Provide fast access to the graph representation of the Software Heritage archive. That endpoint acts as a proxy for the `Software Heritage Graph service `_. It provides fast access to the `graph representation `_ of the Software Heritage archive. For more details please refer to the `Graph RPC API documentation `_. .. warning:: That endpoint is not publicly available and requires authentication and special user permission in order to be able to request it. :param string graph_query: query to forward to the Software Heritage Graph archive (see its `documentation `_) :query boolean resolve_origins: extra parameter defined by that proxy enabling to resolve origin urls from their sha1 representations :statuscode 200: no error :statuscode 400: an invalid graph query has been provided :statuscode 404: provided graph node cannot be found **Examples:** .. parsed-literal:: :swh_web_api:`graph/leaves/swh:1:dir:432d1b21c1256f7408a07c577b6974bbdbcc1323/` :swh_web_api:`graph/neighbors/swh:1:rev:f39d7d78b70e0f39facb1e4fab77ad3df5c52a35/` :swh_web_api:`graph/randomwalk/swh:1:cnt:94a9ed024d3859793618152ea559a168bbcbb5e2/ori?direction=backward` :swh_web_api:`graph/randomwalk/swh:1:cnt:94a9ed024d3859793618152ea559a168bbcbb5e2/ori?direction=backward&limit=-2` :swh_web_api:`graph/visit/nodes/swh:1:snp:40f9f177b8ab0b7b3d70ee14bbc8b214e2b2dcfc?direction=backward&resolve_origins=true` :swh_web_api:`graph/visit/edges/swh:1:snp:40f9f177b8ab0b7b3d70ee14bbc8b214e2b2dcfc?direction=backward&resolve_origins=true` :swh_web_api:`graph/visit/paths/swh:1:dir:644dd466d8ad527ea3a609bfd588a3244e6dafcb?direction=backward&resolve_origins=true` """ return None @api_route(r"/graph/(?P.+)/", "api-1-graph") @renderer_classes([PlainTextRenderer]) def api_graph_proxy( request: Request, graph_query: str ) -> Union[Response, StreamingHttpResponse]: if request.get_host() != SWH_WEB_INTERNAL_SERVER_NAME: if not bool(request.user and request.user.is_authenticated): return Response("Authentication credentials were not provided.", status=401) if not request.user.has_perm(API_GRAPH_PERM): return Response( "You do not have permission to perform this action.", status=403 ) graph_query_url = get_config()["graph"]["server_url"] graph_query_url += graph_query if request.GET: graph_query_url += "?" + request.GET.urlencode(safe="/;:") response = requests.get(graph_query_url, stream=True) # graph stats and counter endpoint responses are not streamed if response.headers.get("Transfer-Encoding") != "chunked": return Response( response.text, status=response.status_code, content_type=response.headers["Content-Type"], ) # other endpoint responses are streamed else: resolve_origins = strtobool(request.GET.get("resolve_origins", "false")) if response.status_code == 200 and resolve_origins: response_stream = _resolve_origin_swhids_in_graph_response(response) else: response_stream = map(lambda line: line + b"\n", response.iter_lines()) return StreamingHttpResponse( response_stream, status=response.status_code, content_type=response.headers["Content-Type"], ) diff --git a/swh/web/api/views/release.py b/swh/web/api/views/release.py index a41b51c8..36d0a8a9 100644 --- a/swh/web/api/views/release.py +++ b/swh/web/api/views/release.py @@ -1,60 +1,60 @@ # Copyright (C) 2015-2019 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from swh.web.api import utils from swh.web.api.apidoc import api_doc, format_docstring from swh.web.api.apiurls import api_route from swh.web.api.views.utils import api_lookup from swh.web.common import archive @api_route( r"/release/(?P[0-9a-f]+)/", "api-1-release", checksum_args=["sha1_git"] ) @api_doc("/release/") @format_docstring() def api_release(request, sha1_git): """ .. http:get:: /api/1/release/(sha1_git)/ Get information about a release in the archive. Releases are identified by **sha1** checksums, compatible with Git tag identifiers. See - :func:`swh.model.identifiers.release_identifier` in our data model + :func:`swh.model.git_objects.release_git_object` in our data model module for details about how they are computed. :param string sha1_git: hexadecimal representation of the release **sha1_git** identifier {common_headers} :>json object author: information about the author of the release :>json string date: ISO representation of the release date (in UTC) :>json string id: the release unique identifier :>json string message: the message associated to the release :>json string name: the name of the release :>json string target: the target identifier of the release :>json string target_type: the type of the target, can be either **release**, **revision**, **content**, **directory** :>json string target_url: a link to the adequate api url based on the target type :statuscode 200: no error :statuscode 400: an invalid **sha1_git** value has been provided :statuscode 404: requested release can not be found in the archive **Example:** .. parsed-literal:: :swh_web_api:`release/208f61cc7a5dbc9879ae6e5c2f95891e270f09ef/` """ error_msg = "Release with sha1_git %s not found." % sha1_git return api_lookup( archive.lookup_release, sha1_git, notfound_msg=error_msg, enrich_fn=utils.enrich_release, request=request, ) diff --git a/swh/web/api/views/snapshot.py b/swh/web/api/views/snapshot.py index dd2bd9db..baca72bb 100644 --- a/swh/web/api/views/snapshot.py +++ b/swh/web/api/views/snapshot.py @@ -1,104 +1,104 @@ # Copyright (C) 2018-2019 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from swh.web.api.apidoc import api_doc, format_docstring from swh.web.api.apiurls import api_route from swh.web.api.utils import enrich_snapshot from swh.web.api.views.utils import api_lookup from swh.web.common import archive from swh.web.common.utils import reverse from swh.web.config import get_config @api_route( r"/snapshot/(?P[0-9a-f]+)/", "api-1-snapshot", checksum_args=["snapshot_id"], ) @api_doc("/snapshot/") @format_docstring() def api_snapshot(request, snapshot_id): """ .. http:get:: /api/1/snapshot/(snapshot_id)/ Get information about a snapshot in the archive. A snapshot is a set of named branches, which are pointers to objects at any level of the Software Heritage DAG. It represents a full picture of an origin at a given time. As well as pointing to other objects in the Software Heritage DAG, branches can also be aliases, in which case their target is the name of another branch in the same snapshot, or dangling, in which case the target is unknown. A snapshot identifier is a salted sha1. See - :func:`swh.model.identifiers.snapshot_identifier` in our data model + :func:`swh.model.git_objects.snapshot_git_object` in our data model module for details about how they are computed. :param sha1 snapshot_id: a snapshot identifier :query str branches_from: optional parameter used to skip branches whose name is lesser than it before returning them :query int branches_count: optional parameter used to restrain the amount of returned branches (default to 1000) :query str target_types: optional comma separated list parameter used to filter the target types of branch to return (possible values that can be contained in that list are ``content``, ``directory``, ``revision``, ``release``, ``snapshot`` or ``alias``) {common_headers} {resheader_link} :>json object branches: object containing all branches associated to the snapshot,for each of them the associated target type and id are given but also a link to get information about that target :>json string id: the unique identifier of the snapshot :statuscode 200: no error :statuscode 400: an invalid snapshot identifier has been provided :statuscode 404: requested snapshot can not be found in the archive **Example:** .. parsed-literal:: :swh_web_api:`snapshot/6a3a2cf0b2b90ce7ae1cf0a221ed68035b686f5a/` """ snapshot_content_max_size = get_config()["snapshot_content_max_size"] branches_from = request.GET.get("branches_from", "") branches_count = int(request.GET.get("branches_count", snapshot_content_max_size)) target_types = request.GET.get("target_types", None) target_types = target_types.split(",") if target_types else None results = api_lookup( archive.lookup_snapshot, snapshot_id, branches_from, branches_count, target_types, branch_name_exclude_prefix=None, notfound_msg="Snapshot with id {} not found.".format(snapshot_id), enrich_fn=enrich_snapshot, request=request, ) response = {"results": results, "headers": {}} if results["next_branch"] is not None: response["headers"]["link-next"] = reverse( "api-1-snapshot", url_args={"snapshot_id": snapshot_id}, query_params={ "branches_from": results["next_branch"], "branches_count": branches_count, "target_types": target_types, }, request=request, ) return response diff --git a/swh/web/api/views/vault.py b/swh/web/api/views/vault.py index 577f8814..18549e4c 100644 --- a/swh/web/api/views/vault.py +++ b/swh/web/api/views/vault.py @@ -1,505 +1,505 @@ # Copyright (C) 2015-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from typing import Any, Dict from django.http import HttpResponse from django.shortcuts import redirect from swh.model.hashutil import hash_to_hex -from swh.model.identifiers import CoreSWHID, ObjectType +from swh.model.swhids import CoreSWHID, ObjectType from swh.web.api.apidoc import api_doc, format_docstring from swh.web.api.apiurls import api_route from swh.web.api.views.utils import api_lookup from swh.web.common import archive, query from swh.web.common.exc import BadInputExc from swh.web.common.utils import reverse ###################################################### # Common SWHID_RE = "swh:1:[a-z]{3}:[0-9a-z]{40}" # XXX: a bit spaghetti. Would be better with class-based views. def _dispatch_cook_progress(request, bundle_type: str, swhid: CoreSWHID): if request.method == "GET": return api_lookup( archive.vault_progress, bundle_type, swhid, notfound_msg=f"Cooking of {swhid} was never requested.", request=request, ) elif request.method == "POST": email = request.POST.get("email", request.GET.get("email", None)) return api_lookup( archive.vault_cook, bundle_type, swhid, email, notfound_msg=f"{swhid} not found.", request=request, ) def _vault_response( vault_response: Dict[str, Any], add_legacy_items: bool ) -> Dict[str, Any]: d = { "fetch_url": vault_response["fetch_url"], "progress_message": vault_response["progress_msg"], "id": vault_response["task_id"], "status": vault_response["task_status"], "swhid": str(vault_response["swhid"]), } if add_legacy_items: d["obj_type"] = vault_response["swhid"].object_type.name.lower() d["obj_id"] = hash_to_hex(vault_response["swhid"].object_id) return d ###################################################### # Flat bundles @api_route( f"/vault/flat/(?P{SWHID_RE})/", "api-1-vault-cook-flat", methods=["GET", "POST"], throttle_scope="swh_vault_cooking", never_cache=True, ) @api_doc("/vault/flat/") @format_docstring() def api_vault_cook_flat(request, swhid): """ .. http:get:: /api/1/vault/flat/(swhid)/ .. http:post:: /api/1/vault/flat/(swhid)/ Request the cooking of a simple archive, typically for a directory. That endpoint enables to create a vault cooking task for a directory through a POST request or check the status of a previously created one through a GET request. Once the cooking task has been executed, the resulting archive can be downloaded using the dedicated endpoint :http:get:`/api/1/vault/flat/(swhid)/raw/`. Then to extract the cooked directory in the current one, use:: $ tar xvf path/to/swh_1_*.tar.gz :param string swhid: the object's SWHID :query string email: e-mail to notify when the archive is ready {common_headers} :>json string fetch_url: the url from which to download the archive once it has been cooked (see :http:get:`/api/1/vault/flat/(swhid)/raw/`) :>json string progress_message: message describing the cooking task progress :>json number id: the cooking task id :>json string status: the cooking task status (either **new**, **pending**, **done** or **failed**) :>json string swhid: the identifier of the object to cook :statuscode 200: no error :statuscode 400: an invalid directory identifier has been provided :statuscode 404: requested directory did not receive any cooking request yet (in case of GET) or can not be found in the archive (in case of POST) """ swhid = CoreSWHID.from_string(swhid) if swhid.object_type == ObjectType.DIRECTORY: res = _dispatch_cook_progress(request, "flat", swhid) res["fetch_url"] = reverse( "api-1-vault-fetch-flat", url_args={"swhid": str(swhid)}, request=request, ) return _vault_response(res, add_legacy_items=False) elif swhid.object_type == ObjectType.CONTENT: raise BadInputExc( "Content objects do not need to be cooked, " "use `/api/1/content/raw/` instead." ) elif swhid.object_type == ObjectType.REVISION: # TODO: support revisions too? (the vault allows it) raise BadInputExc( "Only directories can be cooked as 'flat' bundles. " "Use `/api/1/vault/gitfast/` to cook revisions, as gitfast bundles." ) else: raise BadInputExc("Only directories can be cooked as 'flat' bundles.") @api_route( r"/vault/directory/(?P[0-9a-f]+)/", "api-1-vault-cook-directory", methods=["GET", "POST"], checksum_args=["dir_id"], throttle_scope="swh_vault_cooking", never_cache=True, ) @api_doc("/vault/directory/", tags=["deprecated"]) @format_docstring() def api_vault_cook_directory(request, dir_id): """ .. http:get:: /api/1/vault/directory/(dir_id)/ This endpoint was replaced by :http:get:`/api/1/vault/flat/(swhid)/` """ _, obj_id = query.parse_hash_with_algorithms_or_throws( dir_id, ["sha1"], "Only sha1_git is supported." ) swhid = f"swh:1:dir:{obj_id.hex()}" res = _dispatch_cook_progress(request, "flat", CoreSWHID.from_string(swhid)) res["fetch_url"] = reverse( "api-1-vault-fetch-flat", url_args={"swhid": swhid}, request=request, ) return _vault_response(res, add_legacy_items=True) @api_route( f"/vault/flat/(?P{SWHID_RE})/raw/", "api-1-vault-fetch-flat", ) @api_doc("/vault/flat/raw/") def api_vault_fetch_flat(request, swhid): """ .. http:get:: /api/1/vault/flat/(swhid)/raw/ Fetch the cooked archive for a flat bundle. See :http:get:`/api/1/vault/flat/(swhid)/` to get more details on 'flat' bundle cooking. :param string swhid: the SWHID of the object to cook :resheader Content-Type: application/gzip :statuscode 200: no error :statuscode 404: requested directory did not receive any cooking request yet (in case of GET) or can not be found in the archive (in case of POST) """ res = api_lookup( archive.vault_fetch, "flat", CoreSWHID.from_string(swhid), notfound_msg=f"Cooked archive for {swhid} not found.", request=request, ) fname = "{}.tar.gz".format(swhid) response = HttpResponse(res, content_type="application/gzip") response["Content-disposition"] = "attachment; filename={}".format( fname.replace(":", "_") ) return response @api_route( r"/vault/directory/(?P[0-9a-f]+)/raw/", "api-1-vault-fetch-directory", checksum_args=["dir_id"], ) @api_doc("/vault/directory/raw/", tags=["hidden", "deprecated"]) def api_vault_fetch_directory(request, dir_id): """ .. http:get:: /api/1/vault/directory/(dir_id)/raw/ This endpoint was replaced by :http:get:`/api/1/vault/flat/(swhid)/raw/` """ _, obj_id = query.parse_hash_with_algorithms_or_throws( dir_id, ["sha1"], "Only sha1_git is supported." ) rev_flat_raw_url = reverse( "api-1-vault-fetch-flat", url_args={"swhid": f"swh:1:dir:{dir_id}"} ) return redirect(rev_flat_raw_url) ###################################################### # gitfast bundles @api_route( f"/vault/gitfast/(?P{SWHID_RE})/", "api-1-vault-cook-gitfast", methods=["GET", "POST"], throttle_scope="swh_vault_cooking", never_cache=True, ) @api_doc("/vault/gitfast/") @format_docstring() def api_vault_cook_gitfast(request, swhid): """ .. http:get:: /api/1/vault/gitfast/(swhid)/ .. http:post:: /api/1/vault/gitfast/(swhid)/ Request the cooking of a gitfast archive for a revision or check its cooking status. That endpoint enables to create a vault cooking task for a revision through a POST request or check the status of a previously created one through a GET request. Once the cooking task has been executed, the resulting gitfast archive can be downloaded using the dedicated endpoint :http:get:`/api/1/vault/gitfast/(swhid)/raw/`. Then to import the revision in the current directory, use:: $ git init $ zcat path/to/swh_1_rev_*.gitfast.gz | git fast-import $ git checkout HEAD :param string swhid: the revision's permanent identifiers :query string email: e-mail to notify when the gitfast archive is ready {common_headers} :>json string fetch_url: the url from which to download the archive once it has been cooked (see :http:get:`/api/1/vault/gitfast/(swhid)/raw/`) :>json string progress_message: message describing the cooking task progress :>json number id: the cooking task id :>json string status: the cooking task status (new/pending/done/failed) :>json string swhid: the identifier of the object to cook :statuscode 200: no error :statuscode 404: requested directory did not receive any cooking request yet (in case of GET) or can not be found in the archive (in case of POST) """ swhid = CoreSWHID.from_string(swhid) if swhid.object_type == ObjectType.REVISION: res = _dispatch_cook_progress(request, "gitfast", swhid) res["fetch_url"] = reverse( "api-1-vault-fetch-gitfast", url_args={"swhid": str(swhid)}, request=request, ) return _vault_response(res, add_legacy_items=False) elif swhid.object_type == ObjectType.CONTENT: raise BadInputExc( "Content objects do not need to be cooked, " "use `/api/1/content/raw/` instead." ) elif swhid.object_type == ObjectType.DIRECTORY: raise BadInputExc( "Only revisions can be cooked as 'gitfast' bundles. " "Use `/api/1/vault/flat/` to cook directories, as flat bundles." ) else: raise BadInputExc("Only revisions can be cooked as 'gitfast' bundles.") @api_route( r"/vault/revision/(?P[0-9a-f]+)/gitfast/", "api-1-vault-cook-revision_gitfast", methods=["GET", "POST"], checksum_args=["rev_id"], throttle_scope="swh_vault_cooking", never_cache=True, ) @api_doc("/vault/revision/gitfast/", tags=["deprecated"]) @format_docstring() def api_vault_cook_revision_gitfast(request, rev_id): """ .. http:get:: /api/1/vault/revision/(rev_id)/gitfast/ This endpoint was replaced by :http:get:`/api/1/vault/gitfast/(swhid)/` """ _, obj_id = query.parse_hash_with_algorithms_or_throws( rev_id, ["sha1"], "Only sha1_git is supported." ) swhid = f"swh:1:rev:{obj_id.hex()}" res = _dispatch_cook_progress(request, "gitfast", CoreSWHID.from_string(swhid)) res["fetch_url"] = reverse( "api-1-vault-fetch-gitfast", url_args={"swhid": swhid}, request=request, ) return _vault_response(res, add_legacy_items=True) @api_route( f"/vault/gitfast/(?P{SWHID_RE})/raw/", "api-1-vault-fetch-gitfast", ) @api_doc("/vault/gitfast/raw/") def api_vault_fetch_revision_gitfast(request, swhid): """ .. http:get:: /api/1/vault/gitfast/(swhid)/raw/ Fetch the cooked gitfast archive for a revision. See :http:get:`/api/1/vault/gitfast/(swhid)/` to get more details on gitfast cooking. :param string rev_id: the revision's sha1 identifier :resheader Content-Type: application/gzip :statuscode 200: no error :statuscode 404: requested directory did not receive any cooking request yet (in case of GET) or can not be found in the archive (in case of POST) """ res = api_lookup( archive.vault_fetch, "gitfast", CoreSWHID.from_string(swhid), notfound_msg="Cooked archive for {} not found.".format(swhid), request=request, ) fname = "{}.gitfast.gz".format(swhid) response = HttpResponse(res, content_type="application/gzip") response["Content-disposition"] = "attachment; filename={}".format( fname.replace(":", "_") ) return response @api_route( r"/vault/revision/(?P[0-9a-f]+)/gitfast/raw/", "api-1-vault-fetch-revision_gitfast", checksum_args=["rev_id"], ) @api_doc("/vault/revision_gitfast/raw/", tags=["hidden", "deprecated"]) def _api_vault_revision_gitfast_raw(request, rev_id): """ .. http:get:: /api/1/vault/revision/(rev_id)/gitfast/raw/ This endpoint was replaced by :http:get:`/api/1/vault/gitfast/(swhid)/raw/` """ rev_gitfast_raw_url = reverse( "api-1-vault-fetch-gitfast", url_args={"swhid": f"swh:1:rev:{rev_id}"} ) return redirect(rev_gitfast_raw_url) ###################################################### # git_bare bundles @api_route( f"/vault/git-bare/(?P{SWHID_RE})/", "api-1-vault-cook-git-bare", methods=["GET", "POST"], throttle_scope="swh_vault_cooking", never_cache=True, ) @api_doc("/vault/git-bare/") @format_docstring() def api_vault_cook_git_bare(request, swhid): """ .. http:get:: /api/1/vault/git-bare/(swhid)/ .. http:post:: /api/1/vault/git-bare/(swhid)/ Request the cooking of a git-bare archive for a revision or check its cooking status. That endpoint enables to create a vault cooking task for a revision through a POST request or check the status of a previously created one through a GET request. Once the cooking task has been executed, the resulting git-bare archive can be downloaded using the dedicated endpoint :http:get:`/api/1/vault/git-bare/(swhid)/raw/`. Then to import the revision in the current directory, use:: $ tar -xf path/to/swh_1_rev_*.git.tar $ git clone swh:1:rev:*.git new_repository (replace ``swh:1:rev:*`` with the SWHID of the requested revision) This will create a directory called ``new_repository``, which is a git repository containing the requested objects. :param string swhid: the revision's permanent identifier :query string email: e-mail to notify when the git-bare archive is ready {common_headers} :>json string fetch_url: the url from which to download the archive once it has been cooked (see :http:get:`/api/1/vault/git-bare/(swhid)/raw/`) :>json string progress_message: message describing the cooking task progress :>json number id: the cooking task id :>json string status: the cooking task status (new/pending/done/failed) :>json string swhid: the identifier of the object to cook :statuscode 200: no error :statuscode 404: requested directory did not receive any cooking request yet (in case of GET) or can not be found in the archive (in case of POST) """ swhid = CoreSWHID.from_string(swhid) if swhid.object_type == ObjectType.REVISION: res = _dispatch_cook_progress(request, "git_bare", swhid) res["fetch_url"] = reverse( "api-1-vault-fetch-git-bare", url_args={"swhid": str(swhid)}, request=request, ) return _vault_response(res, add_legacy_items=False) elif swhid.object_type == ObjectType.CONTENT: raise BadInputExc( "Content objects do not need to be cooked, " "use `/api/1/content/raw/` instead." ) elif swhid.object_type == ObjectType.DIRECTORY: raise BadInputExc( "Only revisions can be cooked as 'git-bare' bundles. " "Use `/api/1/vault/flat/` to cook directories, as flat bundles." ) else: raise BadInputExc("Only revisions can be cooked as 'git-bare' bundles.") @api_route( f"/vault/git-bare/(?P{SWHID_RE})/raw/", "api-1-vault-fetch-git-bare", ) @api_doc("/vault/git-bare/raw/") def api_vault_fetch_revision_git_bare(request, swhid): """ .. http:get:: /api/1/vault/git-bare/(swhid)/raw/ Fetch the cooked git-bare archive for a revision. See :http:get:`/api/1/vault/git-bare/(swhid)/` to get more details on git-bare cooking. :param string swhid: the revision's permanent identifier :resheader Content-Type: application/x-tar :statuscode 200: no error :statuscode 404: requested directory did not receive any cooking request yet (in case of GET) or can not be found in the archive (in case of POST) """ res = api_lookup( archive.vault_fetch, "git_bare", CoreSWHID.from_string(swhid), notfound_msg="Cooked archive for {} not found.".format(swhid), request=request, ) fname = "{}.git.tar".format(swhid) response = HttpResponse(res, content_type="application/x-tar") response["Content-disposition"] = "attachment; filename={}".format( fname.replace(":", "_") ) return response diff --git a/swh/web/browse/snapshot_context.py b/swh/web/browse/snapshot_context.py index 4e2f31f2..1439d468 100644 --- a/swh/web/browse/snapshot_context.py +++ b/swh/web/browse/snapshot_context.py @@ -1,1489 +1,1489 @@ # Copyright (C) 2018-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information # Utility module for browsing the archive in a snapshot context. from collections import defaultdict from typing import Any, Dict, List, Optional, Tuple from django.core.cache import cache from django.shortcuts import render from django.utils.html import escape from swh.model.hashutil import hash_to_bytes -from swh.model.identifiers import CoreSWHID, ObjectType from swh.model.model import Snapshot +from swh.model.swhids import CoreSWHID, ObjectType from swh.web.browse.utils import ( content_display_max_size, format_log_entries, gen_content_link, gen_release_link, gen_revision_link, gen_revision_log_link, gen_revision_url, gen_snapshot_link, get_directory_entries, get_readme_to_display, prepare_content_for_display, request_content, ) from swh.web.common import archive, highlightjs from swh.web.common.exc import BadInputExc, NotFoundExc, http_status_code_message from swh.web.common.identifiers import get_swhids_info from swh.web.common.origin_visits import get_origin_visit from swh.web.common.typing import ( ContentMetadata, DirectoryMetadata, OriginInfo, SnapshotBranchInfo, SnapshotContext, SnapshotReleaseInfo, SWHObjectInfo, ) from swh.web.common.utils import ( format_utc_iso_date, gen_path_info, reverse, swh_object_icons, ) from swh.web.config import get_config _empty_snapshot_id = Snapshot(branches={}).id.hex() def _get_branch(branches, branch_name, snapshot_id): """ Utility function to get a specific branch from a branches list. Its purpose is to get the default HEAD branch as some software origin (e.g those with svn type) does not have it. In that latter case, check if there is a master branch instead and returns it. """ filtered_branches = [b for b in branches if b["name"] == branch_name] if filtered_branches: return filtered_branches[0] elif branch_name == "HEAD": filtered_branches = [b for b in branches if b["name"].endswith("master")] if filtered_branches: return filtered_branches[0] elif branches: return branches[0] else: # case where a large branches list has been truncated snp = archive.lookup_snapshot( snapshot_id, branches_from=branch_name, branches_count=1, target_types=["revision", "alias"], # pull request branches must be browsable even if they are hidden # by default in branches list branch_name_exclude_prefix=None, ) snp_branch, _, _ = process_snapshot_branches(snp) if snp_branch and snp_branch[0]["name"] == branch_name: branches.append(snp_branch[0]) return snp_branch[0] def _get_release(releases, release_name, snapshot_id): """ Utility function to get a specific release from a releases list. Returns None if the release can not be found in the list. """ filtered_releases = [r for r in releases if r["name"] == release_name] if filtered_releases: return filtered_releases[0] else: # case where a large branches list has been truncated try: # git origins have specific branches for releases snp = archive.lookup_snapshot( snapshot_id, branches_from=f"refs/tags/{release_name}", branches_count=1, target_types=["release"], ) except NotFoundExc: snp = archive.lookup_snapshot( snapshot_id, branches_from=release_name, branches_count=1, target_types=["release"], ) _, snp_release, _ = process_snapshot_branches(snp) if snp_release and snp_release[0]["name"] == release_name: releases.append(snp_release[0]) return snp_release[0] def _branch_not_found( branch_type, branch, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id ): """ Utility function to raise an exception when a specified branch/release can not be found. """ if branch_type == "branch": branch_type = "Branch" branch_type_plural = "branches" target_type = "revision" else: branch_type = "Release" branch_type_plural = "releases" target_type = "release" if snapshot_id and snapshot_sizes[target_type] == 0: msg = "Snapshot with id %s has an empty list" " of %s!" % ( snapshot_id, branch_type_plural, ) elif snapshot_id: msg = "%s %s for snapshot with id %s" " not found!" % ( branch_type, branch, snapshot_id, ) elif visit_id and snapshot_sizes[target_type] == 0: msg = ( "Origin with url %s" " for visit with id %s has an empty list" " of %s!" % (origin_info["url"], visit_id, branch_type_plural) ) elif visit_id: msg = ( "%s %s associated to visit with" " id %s for origin with url %s" " not found!" % (branch_type, branch, visit_id, origin_info["url"]) ) elif snapshot_sizes[target_type] == 0: msg = ( "Origin with url %s" " for visit with timestamp %s has an empty list" " of %s!" % (origin_info["url"], timestamp, branch_type_plural) ) else: msg = ( "%s %s associated to visit with" " timestamp %s for origin with " "url %s not found!" % (branch_type, branch, timestamp, origin_info["url"]) ) raise NotFoundExc(escape(msg)) def process_snapshot_branches( snapshot: Dict[str, Any] ) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo], Dict[str, Any]]: """ Process a dictionary describing snapshot branches: extract those targeting revisions and releases, put them in two different lists, then sort those lists in lexicographical order of the branches' names. Args: snapshot: A dict describing a snapshot as returned for instance by :func:`swh.web.common.archive.lookup_snapshot` Returns: A tuple whose first member is the sorted list of branches targeting revisions, second member the sorted list of branches targeting releases and third member a dict mapping resolved branch aliases to their real target. """ snapshot_branches = snapshot["branches"] branches: Dict[str, SnapshotBranchInfo] = {} branch_aliases: Dict[str, str] = {} releases: Dict[str, SnapshotReleaseInfo] = {} revision_to_branch = defaultdict(set) revision_to_release = defaultdict(set) release_to_branch = defaultdict(set) for branch_name, target in snapshot_branches.items(): if not target: # FIXME: display branches with an unknown target anyway continue target_id = target["target"] target_type = target["target_type"] if target_type == "revision": branches[branch_name] = SnapshotBranchInfo( name=branch_name, alias=False, revision=target_id, date=None, directory=None, message=None, url=None, ) revision_to_branch[target_id].add(branch_name) elif target_type == "release": release_to_branch[target_id].add(branch_name) elif target_type == "alias": branch_aliases[branch_name] = target_id # FIXME: handle pointers to other object types def _add_release_info(branch, release, alias=False): releases[branch] = SnapshotReleaseInfo( name=release["name"], alias=alias, branch_name=branch, date=format_utc_iso_date(release["date"]), directory=None, id=release["id"], message=release["message"], target_type=release["target_type"], target=release["target"], url=None, ) def _add_branch_info(branch, revision, alias=False): branches[branch] = SnapshotBranchInfo( name=branch, alias=alias, revision=revision["id"], directory=revision["directory"], date=format_utc_iso_date(revision["date"]), message=revision["message"], url=None, ) releases_info = archive.lookup_release_multiple(release_to_branch.keys()) for release in releases_info: if release is None: continue branches_to_update = release_to_branch[release["id"]] for branch in branches_to_update: _add_release_info(branch, release) if release["target_type"] == "revision": revision_to_release[release["target"]].update(branches_to_update) revisions = archive.lookup_revision_multiple( set(revision_to_branch.keys()) | set(revision_to_release.keys()) ) for revision in revisions: if not revision: continue for branch in revision_to_branch[revision["id"]]: _add_branch_info(branch, revision) for release_id in revision_to_release[revision["id"]]: releases[release_id]["directory"] = revision["directory"] resolved_aliases = {} for branch_alias, branch_target in branch_aliases.items(): resolved_alias = archive.lookup_snapshot_alias(snapshot["id"], branch_alias) resolved_aliases[branch_alias] = resolved_alias if resolved_alias is None: continue target_type = resolved_alias["target_type"] target = resolved_alias["target"] if target_type == "revision": revision = archive.lookup_revision(target) _add_branch_info(branch_alias, revision, alias=True) elif target_type == "release": release = archive.lookup_release(target) _add_release_info(branch_alias, release, alias=True) if branch_alias in branches: branches[branch_alias]["name"] = branch_alias ret_branches = list(sorted(branches.values(), key=lambda b: b["name"])) ret_releases = list(sorted(releases.values(), key=lambda b: b["name"])) return ret_branches, ret_releases, resolved_aliases def get_snapshot_content( snapshot_id: str, ) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo], Dict[str, Any]]: """Returns the lists of branches and releases associated to a swh snapshot. That list is put in cache in order to speedup the navigation in the swh-web/browse ui. .. warning:: At most 1000 branches contained in the snapshot will be returned for performance reasons. Args: snapshot_id: hexadecimal representation of the snapshot identifier Returns: A tuple with three members. The first one is a list of dict describing the snapshot branches. The second one is a list of dict describing the snapshot releases. The third one is a dict mapping resolved branch aliases to their real target. Raises: NotFoundExc if the snapshot does not exist """ cache_entry_id = "swh_snapshot_%s" % snapshot_id cache_entry = cache.get(cache_entry_id) if cache_entry: return ( cache_entry["branches"], cache_entry["releases"], cache_entry.get("aliases", {}), ) branches: List[SnapshotBranchInfo] = [] releases: List[SnapshotReleaseInfo] = [] aliases: Dict[str, Any] = {} snapshot_content_max_size = get_config()["snapshot_content_max_size"] if snapshot_id: snapshot = archive.lookup_snapshot( snapshot_id, branches_count=snapshot_content_max_size ) branches, releases, aliases = process_snapshot_branches(snapshot) cache.set( cache_entry_id, {"branches": branches, "releases": releases, "aliases": aliases} ) return branches, releases, aliases def get_origin_visit_snapshot( origin_info: OriginInfo, visit_ts: Optional[str] = None, visit_id: Optional[int] = None, snapshot_id: Optional[str] = None, ) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo], Dict[str, Any]]: """Returns the lists of branches and releases associated to an origin for a given visit. The visit is expressed by either: * a snapshot identifier * a timestamp, if no visit with that exact timestamp is found, the closest one from the provided timestamp will be used. If no visit parameter is provided, it returns the list of branches found for the latest visit. That list is put in cache in order to speedup the navigation in the swh-web/browse ui. .. warning:: At most 1000 branches contained in the snapshot will be returned for performance reasons. Args: origin_info: a dict filled with origin information visit_ts: an ISO 8601 datetime string to parse visit_id: visit id for disambiguation in case several visits have the same timestamp snapshot_id: if provided, visit associated to the snapshot will be processed Returns: A tuple with three members. The first one is a list of dict describing the origin branches for the given visit. The second one is a list of dict describing the origin releases for the given visit. The third one is a dict mapping resolved branch aliases to their real target. Raises: NotFoundExc if the origin or its visit are not found """ visit_info = get_origin_visit(origin_info, visit_ts, visit_id, snapshot_id) return get_snapshot_content(visit_info["snapshot"]) def get_snapshot_context( snapshot_id: Optional[str] = None, origin_url: Optional[str] = None, timestamp: Optional[str] = None, visit_id: Optional[int] = None, branch_name: Optional[str] = None, release_name: Optional[str] = None, revision_id: Optional[str] = None, path: Optional[str] = None, browse_context: str = "directory", ) -> SnapshotContext: """ Utility function to compute relevant information when navigating the archive in a snapshot context. The snapshot is either referenced by its id or it will be retrieved from an origin visit. Args: snapshot_id: hexadecimal representation of a snapshot identifier origin_url: an origin_url timestamp: a datetime string for retrieving the closest visit of the origin visit_id: optional visit id for disambiguation in case of several visits with the same timestamp branch_name: optional branch name set when browsing the snapshot in that scope (will default to "HEAD" if not provided) release_name: optional release name set when browsing the snapshot in that scope revision_id: optional revision identifier set when browsing the snapshot in that scope path: optional path of the object currently browsed in the snapshot browse_context: indicates which type of object is currently browsed Returns: A dict filled with snapshot context information. Raises: swh.web.common.exc.NotFoundExc: if no snapshot is found for the visit of an origin. """ assert origin_url is not None or snapshot_id is not None origin_info = None visit_info = None url_args = {} query_params: Dict[str, Any] = {} origin_visits_url = None if origin_url: if visit_id is not None: query_params["visit_id"] = visit_id elif snapshot_id is not None: query_params["snapshot"] = snapshot_id origin_info = archive.lookup_origin({"url": origin_url}) visit_info = get_origin_visit(origin_info, timestamp, visit_id, snapshot_id) formatted_date = format_utc_iso_date(visit_info["date"]) visit_info["formatted_date"] = formatted_date snapshot_id = visit_info["snapshot"] if not snapshot_id: raise NotFoundExc( "No snapshot associated to the visit of origin " "%s on %s" % (escape(origin_url), formatted_date) ) # provided timestamp is not necessarily equals to the one # of the retrieved visit, so get the exact one in order # to use it in the urls generated below if timestamp: timestamp = visit_info["date"] branches, releases, aliases = get_origin_visit_snapshot( origin_info, timestamp, visit_id, snapshot_id ) query_params["origin_url"] = origin_info["url"] origin_visits_url = reverse( "browse-origin-visits", query_params={"origin_url": origin_info["url"]} ) if timestamp is not None: query_params["timestamp"] = format_utc_iso_date( timestamp, "%Y-%m-%dT%H:%M:%SZ" ) visit_url = reverse("browse-origin-directory", query_params=query_params) visit_info["url"] = visit_url branches_url = reverse("browse-origin-branches", query_params=query_params) releases_url = reverse("browse-origin-releases", query_params=query_params) else: assert snapshot_id is not None branches, releases, aliases = get_snapshot_content(snapshot_id) url_args = {"snapshot_id": snapshot_id} branches_url = reverse("browse-snapshot-branches", url_args=url_args) releases_url = reverse("browse-snapshot-releases", url_args=url_args) releases = list(reversed(releases)) snapshot_sizes_cache_id = f"swh_snapshot_{snapshot_id}_sizes" snapshot_sizes = cache.get(snapshot_sizes_cache_id) if snapshot_sizes is None: snapshot_sizes = archive.lookup_snapshot_sizes(snapshot_id) cache.set(snapshot_sizes_cache_id, snapshot_sizes) is_empty = (snapshot_sizes["release"] + snapshot_sizes["revision"]) == 0 swh_snp_id = str( CoreSWHID(object_type=ObjectType.SNAPSHOT, object_id=hash_to_bytes(snapshot_id)) ) if visit_info: timestamp = format_utc_iso_date(visit_info["date"]) if origin_info: browse_view_name = f"browse-origin-{browse_context}" else: browse_view_name = f"browse-snapshot-{browse_context}" release_id = None root_directory = None snapshot_total_size = snapshot_sizes["release"] + snapshot_sizes["revision"] if path is not None: query_params["path"] = path if snapshot_total_size and revision_id is not None: revision = archive.lookup_revision(revision_id) root_directory = revision["directory"] branches.append( SnapshotBranchInfo( name=revision_id, alias=False, revision=revision_id, directory=root_directory, date=revision["date"], message=revision["message"], url=None, ) ) query_params["revision"] = revision_id elif snapshot_total_size and release_name: release = _get_release(releases, release_name, snapshot_id) if release is None: _branch_not_found( "release", release_name, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id, ) else: root_directory = release["directory"] revision_id = release["target"] release_id = release["id"] query_params["release"] = release_name elif snapshot_total_size: if branch_name: query_params["branch"] = branch_name branch = _get_branch(branches, branch_name or "HEAD", snapshot_id) if branch is None: _branch_not_found( "branch", branch_name, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id, ) else: branch_name = branch["name"] revision_id = branch["revision"] root_directory = branch["directory"] for b in branches: branch_query_params = dict(query_params) branch_query_params.pop("release", None) if b["name"] != b["revision"]: branch_query_params.pop("revision", None) branch_query_params["branch"] = b["name"] b["url"] = reverse( browse_view_name, url_args=url_args, query_params=branch_query_params ) for r in releases: release_query_params = dict(query_params) release_query_params.pop("branch", None) release_query_params.pop("revision", None) release_query_params["release"] = r["name"] r["url"] = reverse( browse_view_name, url_args=url_args, query_params=release_query_params, ) revision_info = None if revision_id: try: revision_info = archive.lookup_revision(revision_id) except NotFoundExc: pass else: revision_info["date"] = format_utc_iso_date(revision_info["date"]) revision_info["committer_date"] = format_utc_iso_date( revision_info["committer_date"] ) if revision_info["message"]: message_lines = revision_info["message"].split("\n") revision_info["message_header"] = message_lines[0] else: revision_info["message_header"] = "" snapshot_context = SnapshotContext( branch=branch_name, branch_alias=branch_name in aliases, branches=branches, branches_url=branches_url, is_empty=is_empty, origin_info=origin_info, origin_visits_url=origin_visits_url, release=release_name, release_alias=release_name in aliases, release_id=release_id, query_params=query_params, releases=releases, releases_url=releases_url, revision_id=revision_id, revision_info=revision_info, root_directory=root_directory, snapshot_id=snapshot_id, snapshot_sizes=snapshot_sizes, snapshot_swhid=swh_snp_id, url_args=url_args, visit_info=visit_info, ) if revision_info: revision_info["revision_url"] = gen_revision_url(revision_id, snapshot_context) return snapshot_context def _build_breadcrumbs(snapshot_context: SnapshotContext, path: str): origin_info = snapshot_context["origin_info"] url_args = snapshot_context["url_args"] query_params = dict(snapshot_context["query_params"]) root_directory = snapshot_context["root_directory"] path_info = gen_path_info(path) if origin_info: browse_view_name = "browse-origin-directory" else: browse_view_name = "browse-snapshot-directory" breadcrumbs = [] if root_directory: query_params.pop("path", None) breadcrumbs.append( { "name": root_directory[:7], "url": reverse( browse_view_name, url_args=url_args, query_params=query_params ), } ) for pi in path_info: query_params["path"] = pi["path"] breadcrumbs.append( { "name": pi["name"], "url": reverse( browse_view_name, url_args=url_args, query_params=query_params ), } ) return breadcrumbs def _check_origin_url(snapshot_id, origin_url): if snapshot_id is None and origin_url is None: raise BadInputExc("An origin URL must be provided as query parameter.") def browse_snapshot_directory( request, snapshot_id=None, origin_url=None, timestamp=None, path=None ): """ Django view implementation for browsing a directory in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), path=path, browse_context="directory", branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), ) root_directory = snapshot_context["root_directory"] sha1_git = root_directory error_info = { "status_code": 200, "description": None, } if root_directory and path: try: dir_info = archive.lookup_directory_with_path(root_directory, path) sha1_git = dir_info["target"] except NotFoundExc as e: sha1_git = None error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" dirs = [] files = [] if sha1_git: dirs, files = get_directory_entries(sha1_git) origin_info = snapshot_context["origin_info"] visit_info = snapshot_context["visit_info"] url_args = snapshot_context["url_args"] query_params = dict(snapshot_context["query_params"]) revision_id = snapshot_context["revision_id"] snapshot_id = snapshot_context["snapshot_id"] if origin_info: browse_view_name = "browse-origin-directory" else: browse_view_name = "browse-snapshot-directory" breadcrumbs = _build_breadcrumbs(snapshot_context, path) path = "" if path is None else (path + "/") for d in dirs: if d["type"] == "rev": d["url"] = reverse("browse-revision", url_args={"sha1_git": d["target"]}) else: query_params["path"] = path + d["name"] d["url"] = reverse( browse_view_name, url_args=url_args, query_params=query_params ) sum_file_sizes = 0 readmes = {} if origin_info: browse_view_name = "browse-origin-content" else: browse_view_name = "browse-snapshot-content" for f in files: query_params["path"] = path + f["name"] f["url"] = reverse( browse_view_name, url_args=url_args, query_params=query_params ) if f["length"] is not None: sum_file_sizes += f["length"] if f["name"].lower().startswith("readme"): readmes[f["name"]] = f["checksums"]["sha1"] readme_name, readme_url, readme_html = get_readme_to_display(readmes) if origin_info: browse_view_name = "browse-origin-log" else: browse_view_name = "browse-snapshot-log" history_url = None if snapshot_id != _empty_snapshot_id: query_params.pop("path", None) history_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) nb_files = None nb_dirs = None dir_path = None if root_directory: nb_files = len(files) nb_dirs = len(dirs) dir_path = "/" + path swh_objects = [] vault_cooking = {} revision_found = True if sha1_git is None and revision_id is not None: try: archive.lookup_revision(revision_id) except NotFoundExc: revision_found = False if sha1_git is not None: swh_objects.append( SWHObjectInfo(object_type=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"], ) def browse_snapshot_content( request, snapshot_id=None, origin_url=None, timestamp=None, path=None, selected_language=None, ): """ Django view implementation for browsing a content in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) if path is None: raise BadInputExc("The path of a content must be given as query parameter.") snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), path=path, browse_context="content", branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), ) root_directory = snapshot_context["root_directory"] sha1_git = None query_string = None content_data = {} directory_id = None split_path = path.split("/") filename = split_path[-1] filepath = path[: -len(filename)] error_info = { "status_code": 200, "description": None, } if root_directory: try: content_info = archive.lookup_directory_with_path(root_directory, path) sha1_git = content_info["target"] query_string = "sha1_git:" + sha1_git content_data = request_content(query_string) if filepath: dir_info = archive.lookup_directory_with_path(root_directory, filepath) directory_id = dir_info["target"] else: directory_id = root_directory except NotFoundExc as e: error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" revision_id = snapshot_context["revision_id"] origin_info = snapshot_context["origin_info"] visit_info = snapshot_context["visit_info"] snapshot_id = snapshot_context["snapshot_id"] if content_data.get("raw_data") is not None: content_display_data = prepare_content_for_display( content_data["raw_data"], content_data["mimetype"], path ) content_data.update(content_display_data) # Override language with user-selected language if selected_language is not None: content_data["language"] = selected_language available_languages = None if content_data.get("mimetype") is not None and "text/" in content_data["mimetype"]: available_languages = highlightjs.get_supported_languages() breadcrumbs = _build_breadcrumbs(snapshot_context, filepath) breadcrumbs.append({"name": filename, "url": None}) browse_content_link = gen_content_link(sha1_git) content_raw_url = None if query_string: content_raw_url = reverse( "browse-content-raw", url_args={"query_string": query_string}, query_params={"filename": filename}, ) content_checksums = content_data.get("checksums", {}) sha1_git = content_checksums.get("sha1_git") swh_objects = [] if sha1_git is not None: swh_objects.append( SWHObjectInfo(object_type=ObjectType.CONTENT, object_id=sha1_git) ) if directory_id is not None: swh_objects.append( SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=directory_id) ) if revision_id is not None: swh_objects.append( SWHObjectInfo(object_type=ObjectType.REVISION, object_id=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) ) 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=browse_content_link, mimetype=content_data.get("mimetype"), encoding=content_data.get("encoding"), size=content_data.get("length", 0), language=content_data.get("language"), root_directory=root_directory, path=f"/{filepath}", filename=filename, directory=directory_id, revision=revision_id, release=release_id, snapshot=snapshot_id, origin_url=origin_url, visit_date=visit_date, visit_type=visit_type, ) swhids_info = get_swhids_info(swh_objects, snapshot_context, content_metadata) content_path = "/".join([bc["name"] for bc in breadcrumbs]) context_found = "snapshot: %s" % snapshot_context["snapshot_id"] if origin_info: context_found = "origin: %s" % origin_info["url"] heading = "Content - %s - %s - %s" % ( content_path, snapshot_context["branch"], context_found, ) top_right_link = None if not snapshot_context["is_empty"]: top_right_link = { "url": content_raw_url, "icon": swh_object_icons["content"], "text": "Raw File", } return render( request, "browse/content.html", { "heading": heading, "swh_object_name": "Content", "swh_object_metadata": content_metadata, "content": content_data.get("content_data"), "content_size": content_data.get("length"), "max_content_size": content_display_max_size, "filename": filename, "encoding": content_data.get("encoding"), "mimetype": content_data.get("mimetype"), "language": content_data.get("language"), "available_languages": available_languages, "breadcrumbs": breadcrumbs if root_directory else [], "top_right_link": top_right_link, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": True, "swhids_info": swhids_info, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], }, status=error_info["status_code"], ) PER_PAGE = 100 def browse_snapshot_log(request, snapshot_id=None, origin_url=None, timestamp=None): """ Django view implementation for browsing a revision history in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), browse_context="log", branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), ) revision_id = snapshot_context["revision_id"] per_page = int(request.GET.get("per_page", PER_PAGE)) offset = int(request.GET.get("offset", 0)) revs_ordering = request.GET.get("revs_ordering", "committer_date") session_key = "rev_%s_log_ordering_%s" % (revision_id, revs_ordering) rev_log_session = request.session.get(session_key, None) rev_log = [] revs_walker_state = None if rev_log_session: rev_log = rev_log_session["rev_log"] revs_walker_state = rev_log_session["revs_walker_state"] if len(rev_log) < offset + per_page: revs_walker = archive.get_revisions_walker( revs_ordering, revision_id, max_revs=offset + per_page + 1, state=revs_walker_state, ) rev_log += [rev["id"] for rev in revs_walker] revs_walker_state = revs_walker.export_state() revs = rev_log[offset : offset + per_page] revision_log = archive.lookup_revision_multiple(revs) request.session[session_key] = { "rev_log": rev_log, "revs_walker_state": revs_walker_state, } origin_info = snapshot_context["origin_info"] visit_info = snapshot_context["visit_info"] url_args = snapshot_context["url_args"] query_params = snapshot_context["query_params"] snapshot_id = snapshot_context["snapshot_id"] query_params["per_page"] = per_page revs_ordering = request.GET.get("revs_ordering", "") query_params["revs_ordering"] = revs_ordering or None if origin_info: browse_view_name = "browse-origin-log" else: browse_view_name = "browse-snapshot-log" prev_log_url = None if len(rev_log) > offset + per_page: query_params["offset"] = offset + per_page prev_log_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) next_log_url = None if offset != 0: query_params["offset"] = offset - per_page next_log_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) revision_log_data = format_log_entries(revision_log, per_page, snapshot_context) browse_rev_link = gen_revision_link(revision_id) browse_log_link = gen_revision_log_link(revision_id) browse_snp_link = gen_snapshot_link(snapshot_id) revision_metadata = { "context-independent revision": browse_rev_link, "context-independent revision history": browse_log_link, "context-independent snapshot": browse_snp_link, "snapshot": snapshot_id, } if origin_info: revision_metadata["origin url"] = origin_info["url"] revision_metadata["origin visit date"] = format_utc_iso_date(visit_info["date"]) revision_metadata["origin visit type"] = visit_info["type"] swh_objects = [ SWHObjectInfo(object_type=ObjectType.REVISION, object_id=revision_id), SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id), ] release_id = snapshot_context["release_id"] if release_id: swh_objects.append( SWHObjectInfo(object_type=ObjectType.RELEASE, object_id=release_id) ) browse_rel_link = gen_release_link(release_id) revision_metadata["release"] = release_id revision_metadata["context-independent release"] = browse_rel_link swhids_info = get_swhids_info(swh_objects, snapshot_context) context_found = "snapshot: %s" % snapshot_context["snapshot_id"] if origin_info: context_found = "origin: %s" % origin_info["url"] heading = "Revision history - %s - %s" % (snapshot_context["branch"], context_found) return render( request, "browse/revision-log.html", { "heading": heading, "swh_object_name": "Revisions history", "swh_object_metadata": revision_metadata, "revision_log": revision_log_data, "revs_ordering": revs_ordering, "next_log_url": next_log_url, "prev_log_url": prev_log_url, "breadcrumbs": None, "top_right_link": None, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": True, "swhids_info": swhids_info, }, ) def browse_snapshot_branches( request, snapshot_id=None, origin_url=None, timestamp=None, branch_name_include=None ): """ Django view implementation for browsing a list of branches in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), ) branches_bc = request.GET.get("branches_breadcrumbs", "") branches_bc = branches_bc.split(",") if branches_bc else [] branches_from = branches_bc[-1] if branches_bc else "" origin_info = snapshot_context["origin_info"] url_args = snapshot_context["url_args"] query_params = snapshot_context["query_params"] if origin_info: browse_view_name = "browse-origin-directory" else: browse_view_name = "browse-snapshot-directory" snapshot = archive.lookup_snapshot( snapshot_context["snapshot_id"], branches_from, PER_PAGE + 1, target_types=["revision", "alias"], branch_name_include_substring=branch_name_include, ) displayed_branches = [] if snapshot: displayed_branches, _, _ = process_snapshot_branches(snapshot) for branch in displayed_branches: rev_query_params = {} if origin_info: rev_query_params["origin_url"] = origin_info["url"] revision_url = reverse( "browse-revision", url_args={"sha1_git": branch["revision"]}, query_params=query_params, ) query_params["branch"] = branch["name"] directory_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) del query_params["branch"] branch["revision_url"] = revision_url branch["directory_url"] = directory_url if origin_info: browse_view_name = "browse-origin-branches" else: browse_view_name = "browse-snapshot-branches" prev_branches_url = None next_branches_url = None if branches_bc: query_params_prev = dict(query_params) query_params_prev["branches_breadcrumbs"] = ",".join(branches_bc[:-1]) prev_branches_url = reverse( browse_view_name, url_args=url_args, query_params=query_params_prev ) elif branches_from: prev_branches_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) if snapshot and snapshot["next_branch"] is not None: query_params_next = dict(query_params) next_branch = displayed_branches[-1]["name"] del displayed_branches[-1] branches_bc.append(next_branch) query_params_next["branches_breadcrumbs"] = ",".join(branches_bc) next_branches_url = reverse( browse_view_name, url_args=url_args, query_params=query_params_next ) heading = "Branches - " if origin_info: heading += "origin: %s" % origin_info["url"] else: heading += "snapshot: %s" % snapshot_id return render( request, "browse/branches.html", { "heading": heading, "swh_object_name": "Branches", "swh_object_metadata": {}, "top_right_link": None, "displayed_branches": displayed_branches, "prev_branches_url": prev_branches_url, "next_branches_url": next_branches_url, "snapshot_context": snapshot_context, "search_string": branch_name_include or "", }, ) def browse_snapshot_releases( request, snapshot_id=None, origin_url=None, timestamp=None, release_name_include=None, ): """ Django view implementation for browsing a list of releases in a snapshot context. """ _check_origin_url(snapshot_id, origin_url) snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=request.GET.get("visit_id"), ) rel_bc = request.GET.get("releases_breadcrumbs", "") rel_bc = rel_bc.split(",") if rel_bc else [] rel_from = rel_bc[-1] if rel_bc else "" origin_info = snapshot_context["origin_info"] url_args = snapshot_context["url_args"] query_params = snapshot_context["query_params"] snapshot = archive.lookup_snapshot( snapshot_context["snapshot_id"], rel_from, PER_PAGE + 1, target_types=["release", "alias"], branch_name_include_substring=release_name_include, ) displayed_releases = [] if snapshot: _, displayed_releases, _ = process_snapshot_branches(snapshot) for release in displayed_releases: query_params_tgt = {"snapshot": snapshot_id} if origin_info: query_params_tgt["origin_url"] = origin_info["url"] release_url = reverse( "browse-release", url_args={"sha1_git": release["id"]}, query_params=query_params_tgt, ) target_url = "" tooltip = ( f"The release {release['name']} targets " f"{release['target_type']} {release['target']}" ) if release["target_type"] == "revision": target_url = reverse( "browse-revision", url_args={"sha1_git": release["target"]}, query_params=query_params_tgt, ) elif release["target_type"] == "directory": target_url = reverse( "browse-directory", url_args={"sha1_git": release["target"]}, query_params=query_params_tgt, ) elif release["target_type"] == "content": target_url = reverse( "browse-content", url_args={"query_string": release["target"]}, query_params=query_params_tgt, ) elif release["target_type"] == "release": target_url = reverse( "browse-release", url_args={"sha1_git": release["target"]}, query_params=query_params_tgt, ) tooltip = ( f"The release {release['name']} " f"is an alias for release {release['target']}" ) release["release_url"] = release_url release["target_url"] = target_url release["tooltip"] = tooltip if origin_info: browse_view_name = "browse-origin-releases" else: browse_view_name = "browse-snapshot-releases" prev_releases_url = None next_releases_url = None if rel_bc: query_params_prev = dict(query_params) query_params_prev["releases_breadcrumbs"] = ",".join(rel_bc[:-1]) prev_releases_url = reverse( browse_view_name, url_args=url_args, query_params=query_params_prev ) elif rel_from: prev_releases_url = reverse( browse_view_name, url_args=url_args, query_params=query_params ) if snapshot and snapshot["next_branch"] is not None: query_params_next = dict(query_params) next_rel = displayed_releases[-1]["branch_name"] del displayed_releases[-1] rel_bc.append(next_rel) query_params_next["releases_breadcrumbs"] = ",".join(rel_bc) next_releases_url = reverse( browse_view_name, url_args=url_args, query_params=query_params_next ) heading = "Releases - " if origin_info: heading += "origin: %s" % origin_info["url"] else: heading += "snapshot: %s" % snapshot_id return render( request, "browse/releases.html", { "heading": heading, "top_panel_visible": False, "top_panel_collapsible": False, "swh_object_name": "Releases", "swh_object_metadata": {}, "top_right_link": None, "displayed_releases": displayed_releases, "prev_releases_url": prev_releases_url, "next_releases_url": next_releases_url, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": False, "search_string": release_name_include or "", }, ) diff --git a/swh/web/browse/views/content.py b/swh/web/browse/views/content.py index 497eba0f..a338a19f 100644 --- a/swh/web/browse/views/content.py +++ b/swh/web/browse/views/content.py @@ -1,401 +1,401 @@ # Copyright (C) 2017-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import difflib from distutils.util import strtobool import sentry_sdk from django.http import HttpResponse, JsonResponse from django.shortcuts import render from swh.model.hashutil import hash_to_hex -from swh.model.identifiers import ObjectType +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 NotFoundExc, http_status_code_message from swh.web.common.identifiers import get_swhids_info from swh.web.common.typing import ContentMetadata, SWHObjectInfo from swh.web.common.utils import gen_path_info, reverse, swh_object_icons @browse_route( r"content/(?P[0-9a-z_:]*[0-9a-f]+.)/raw/", view_name="browse-content-raw", checksum_args=["query_string"], ) def content_raw(request, query_string): """Django view that produces a raw display of a content identified by its hash value. The url that points to it is :http:get:`/browse/content/[(algo_hash):](hash)/raw/` """ re_encode = bool(strtobool(request.GET.get("re_encode", "false"))) algo, checksum = query.parse_hash(query_string) checksum = hash_to_hex(checksum) content_data = request_content(query_string, max_size=None, re_encode=re_encode) filename = request.GET.get("filename", None) if not filename: filename = "%s_%s" % (algo, checksum) if ( content_data["mimetype"].startswith("text/") or content_data["mimetype"] == "inode/x-empty" ): response = HttpResponse(content_data["raw_data"], content_type="text/plain") response["Content-disposition"] = "filename=%s" % filename else: response = HttpResponse( content_data["raw_data"], content_type="application/octet-stream" ) response["Content-disposition"] = "attachment; filename=%s" % filename return response _auto_diff_size_limit = 20000 @browse_route( r"content/(?P.*)/diff/(?P.*)", view_name="diff-contents", ) def _contents_diff(request, from_query_string, to_query_string): """ Browse endpoint used to compute unified diffs between two contents. Diffs are generated only if the two contents are textual. By default, diffs whose size are greater than 20 kB will not be generated. To force the generation of large diffs, the 'force' boolean query parameter must be used. Args: request: input django http request from_query_string: a string of the form "[ALGO_HASH:]HASH" where optional ALGO_HASH can be either ``sha1``, ``sha1_git``, ``sha256``, or ``blake2s256`` (default to ``sha1``) and HASH the hexadecimal representation of the hash value identifying the first content to_query_string: same as above for identifying the second content Returns: A JSON object containing the unified diff. """ diff_data = {} content_from = None content_to = None content_from_size = 0 content_to_size = 0 content_from_lines = [] content_to_lines = [] force = request.GET.get("force", "false") path = request.GET.get("path", None) language = "nohighlight" force = bool(strtobool(force)) if from_query_string == to_query_string: diff_str = "File renamed without changes" else: try: text_diff = True if from_query_string: content_from = request_content(from_query_string, max_size=None) content_from_display_data = prepare_content_for_display( content_from["raw_data"], content_from["mimetype"], path ) language = content_from_display_data["language"] content_from_size = content_from["length"] if not ( content_from["mimetype"].startswith("text/") or content_from["mimetype"] == "inode/x-empty" ): text_diff = False if text_diff and to_query_string: content_to = request_content(to_query_string, max_size=None) content_to_display_data = prepare_content_for_display( content_to["raw_data"], content_to["mimetype"], path ) language = content_to_display_data["language"] content_to_size = content_to["length"] if not ( content_to["mimetype"].startswith("text/") or content_to["mimetype"] == "inode/x-empty" ): text_diff = False diff_size = abs(content_to_size - content_from_size) if not text_diff: diff_str = "Diffs are not generated for non textual content" language = "nohighlight" elif not force and diff_size > _auto_diff_size_limit: diff_str = "Large diffs are not automatically computed" language = "nohighlight" else: if content_from: content_from_lines = ( content_from["raw_data"].decode("utf-8").splitlines(True) ) if content_from_lines and content_from_lines[-1][-1] != "\n": content_from_lines[-1] += "[swh-no-nl-marker]\n" if content_to: content_to_lines = ( content_to["raw_data"].decode("utf-8").splitlines(True) ) if content_to_lines and content_to_lines[-1][-1] != "\n": content_to_lines[-1] += "[swh-no-nl-marker]\n" diff_lines = difflib.unified_diff(content_from_lines, content_to_lines) diff_str = "".join(list(diff_lines)[2:]) except Exception as exc: sentry_sdk.capture_exception(exc) diff_str = str(exc) diff_data["diff_str"] = diff_str diff_data["language"] = language return JsonResponse(diff_data) @browse_route( r"content/(?P[0-9a-z_:]*[0-9a-f]+.)/", view_name="browse-content", checksum_args=["query_string"], ) def content_display(request, query_string): """Django view that produces an HTML display of a content identified by its hash value. The url that points to it is :http:get:`/browse/content/[(algo_hash):](hash)/` """ algo, checksum = query.parse_hash(query_string) checksum = hash_to_hex(checksum) origin_url = request.GET.get("origin_url") selected_language = request.GET.get("language") if not origin_url: origin_url = request.GET.get("origin") snapshot_id = request.GET.get("snapshot") path = request.GET.get("path") content_data = {} error_info = {"status_code": 200, "description": None} try: content_data = request_content(query_string) except NotFoundExc as e: error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" snapshot_context = None if origin_url is not None or snapshot_id is not None: try: snapshot_context = get_snapshot_context( origin_url=origin_url, snapshot_id=snapshot_id, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), path=path, browse_context="content", ) except NotFoundExc as e: if str(e).startswith("Origin"): raw_cnt_url = reverse( "browse-content", url_args={"query_string": query_string} ) error_message = ( "The Software Heritage archive has a content " "with the hash you provided but the origin " "mentioned in your request appears broken: %s. " "Please check the URL and try again.\n\n" "Nevertheless, you can still browse the content " "without origin information: %s" % (gen_link(origin_url), gen_link(raw_cnt_url)) ) raise NotFoundExc(error_message) else: raise e content = None language = None mimetype = None if content_data.get("raw_data") is not None: content_display_data = prepare_content_for_display( content_data["raw_data"], content_data["mimetype"], path ) content = content_display_data["content_data"] language = content_display_data["language"] mimetype = content_display_data["mimetype"] # Override language with user-selected language if selected_language is not None: language = selected_language available_languages = None if mimetype and "text/" in mimetype: available_languages = highlightjs.get_supported_languages() filename = None path_info = None directory_id = None root_dir = None if snapshot_context: root_dir = snapshot_context.get("root_directory") query_params = snapshot_context["query_params"] if snapshot_context else {} breadcrumbs = [] if path: split_path = path.split("/") root_dir = root_dir or split_path[0] filename = split_path[-1] if root_dir != path: path = path.replace(root_dir + "/", "") path = path[: -len(filename)] path_info = gen_path_info(path) query_params.pop("path", None) dir_url = reverse( "browse-directory", url_args={"sha1_git": root_dir}, query_params=query_params, ) breadcrumbs.append({"name": root_dir[:7], "url": dir_url}) for pi in path_info: query_params["path"] = pi["path"] dir_url = reverse( "browse-directory", url_args={"sha1_git": root_dir}, query_params=query_params, ) breadcrumbs.append({"name": pi["name"], "url": dir_url}) breadcrumbs.append({"name": filename, "url": None}) if path and root_dir != path: dir_info = archive.lookup_directory_with_path(root_dir, path) directory_id = dir_info["target"] elif root_dir != path: directory_id = root_dir else: root_dir = None query_params = {"filename": filename} content_checksums = content_data.get("checksums", {}) content_url = reverse("browse-content", url_args={"query_string": query_string},) content_raw_url = reverse( "browse-content-raw", url_args={"query_string": query_string}, query_params=query_params, ) content_metadata = ContentMetadata( object_type=ObjectType.CONTENT, object_id=content_checksums.get("sha1_git"), sha1=content_checksums.get("sha1"), sha1_git=content_checksums.get("sha1_git"), sha256=content_checksums.get("sha256"), blake2s256=content_checksums.get("blake2s256"), content_url=content_url, mimetype=content_data.get("mimetype"), encoding=content_data.get("encoding"), size=content_data.get("length", 0), language=content_data.get("language"), root_directory=root_dir, path=f"/{path}" if path else None, filename=filename or "", directory=directory_id, revision=None, release=None, snapshot=None, origin_url=origin_url, ) swh_objects = [ 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: swh_objects.append( SWHObjectInfo( object_type=ObjectType.REVISION, object_id=snapshot_context["revision_id"], ) ) swh_objects.append( SWHObjectInfo( object_type=ObjectType.SNAPSHOT, object_id=snapshot_context["snapshot_id"], ) ) if snapshot_context["release_id"]: swh_objects.append( SWHObjectInfo( object_type=ObjectType.RELEASE, object_id=snapshot_context["release_id"], ) ) swhids_info = get_swhids_info( swh_objects, snapshot_context, extra_context=content_metadata, ) heading = "Content - %s" % content_checksums.get("sha1_git") if breadcrumbs: content_path = "/".join([bc["name"] for bc in breadcrumbs]) heading += " - %s" % content_path return render( request, "browse/content.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"], "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 56b7654e..22428510 100644 --- a/swh/web/browse/views/directory.py +++ b/swh/web/browse/views/directory.py @@ -1,281 +1,281 @@ # Copyright (C) 2017-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import os import sentry_sdk from django.http import HttpResponse from django.shortcuts import redirect, render -from swh.model.identifiers import ObjectType +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 from swh.web.common.identifiers import get_swhids_info from swh.web.common.typing import DirectoryMetadata, SWHObjectInfo from swh.web.common.utils import gen_path_info, reverse, swh_object_icons def _directory_browse(request, sha1_git, path=None): root_sha1_git = sha1_git error_info = {"status_code": 200, "description": None} if path: try: dir_info = archive.lookup_directory_with_path(sha1_git, path) sha1_git = dir_info["target"] except NotFoundExc as e: error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" sha1_git = None dirs, files = [], [] if sha1_git is not None: dirs, files = get_directory_entries(sha1_git) origin_url = request.GET.get("origin_url") if not origin_url: origin_url = request.GET.get("origin") snapshot_id = request.GET.get("snapshot") snapshot_context = None if origin_url is not None or snapshot_id is not None: try: snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=request.GET.get("revision"), path=path, ) except NotFoundExc as e: if str(e).startswith("Origin"): raw_dir_url = reverse( "browse-directory", url_args={"sha1_git": sha1_git} ) error_message = ( "The Software Heritage archive has a directory " "with the hash you provided but the origin " "mentioned in your request appears broken: %s. " "Please check the URL and try again.\n\n" "Nevertheless, you can still browse the directory " "without origin information: %s" % (gen_link(origin_url), gen_link(raw_dir_url)) ) raise NotFoundExc(error_message) else: raise e path_info = gen_path_info(path) query_params = snapshot_context["query_params"] if snapshot_context else {} breadcrumbs = [] breadcrumbs.append( { "name": root_sha1_git[:7], "url": reverse( "browse-directory", url_args={"sha1_git": root_sha1_git}, query_params={**query_params, "path": None}, ), } ) for pi in path_info: breadcrumbs.append( { "name": pi["name"], "url": reverse( "browse-directory", url_args={"sha1_git": root_sha1_git}, query_params={**query_params, "path": pi["path"],}, ), } ) path = "" if path is None else (path + "/") for d in dirs: if d["type"] == "rev": d["url"] = reverse( "browse-revision", url_args={"sha1_git": d["target"]}, query_params=query_params, ) else: d["url"] = reverse( "browse-directory", url_args={"sha1_git": root_sha1_git}, query_params={**query_params, "path": path + d["name"],}, ) sum_file_sizes = 0 readmes = {} for f in files: query_string = "sha1_git:" + f["target"] f["url"] = reverse( "browse-content", url_args={"query_string": query_string}, query_params={ **query_params, "path": root_sha1_git + "/" + path + f["name"], }, ) if f["length"] is not None: sum_file_sizes += f["length"] if f["name"].lower().startswith("readme"): readmes[f["name"]] = f["checksums"]["sha1"] readme_name, readme_url, readme_html = get_readme_to_display(readmes) dir_metadata = DirectoryMetadata( object_type=ObjectType.DIRECTORY, object_id=sha1_git, directory=root_sha1_git, nb_files=len(files), nb_dirs=len(dirs), sum_file_sizes=sum_file_sizes, root_directory=root_sha1_git, path=f"/{path}" if path else None, revision=None, revision_found=None, release=None, snapshot=None, ) vault_cooking = { "directory_context": True, "directory_swhid": f"swh:1:dir:{sha1_git}", "revision_context": False, "revision_swhid": None, } swh_objects = [SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=sha1_git)] if snapshot_context: 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 if breadcrumbs: dir_path = "/".join([bc["name"] for bc in breadcrumbs]) + "/" heading += " - %s" % dir_path top_right_link = None if snapshot_context is not None and not snapshot_context["is_empty"]: history_url = reverse( "browse-revision-log", url_args={"sha1_git": snapshot_context["revision_id"]}, query_params=query_params, ) top_right_link = { "url": history_url, "icon": swh_object_icons["revisions history"], "text": "History", } return render( request, "browse/directory.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"], "swh_object_name": "Directory", "swh_object_metadata": dir_metadata, "dirs": dirs, "files": files, "breadcrumbs": breadcrumbs, "top_right_link": top_right_link, "readme_name": readme_name, "readme_url": readme_url, "readme_html": readme_html, "snapshot_context": snapshot_context, "vault_cooking": vault_cooking, "show_actions": True, "swhids_info": swhids_info, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], }, status=error_info["status_code"], ) @browse_route( r"directory/(?P[0-9a-f]+)/", view_name="browse-directory", checksum_args=["sha1_git"], ) def directory_browse(request, sha1_git): """Django view for browsing the content of a directory identified by its sha1_git value. The url that points to it is :http:get:`/browse/directory/(sha1_git)/` """ return _directory_browse(request, sha1_git, request.GET.get("path")) @browse_route( r"directory/(?P[0-9a-f]+)/(?P.+)/", view_name="browse-directory-legacy", checksum_args=["sha1_git"], ) def directory_browse_legacy(request, sha1_git, path): """Django view for browsing the content of a directory identified by its sha1_git value. The url that points to it is :http:get:`/browse/directory/(sha1_git)/(path)/` """ return _directory_browse(request, sha1_git, path) @browse_route( r"directory/resolve/content-path/(?P[0-9a-f]+)/", view_name="browse-directory-resolve-content-path", checksum_args=["sha1_git"], ) def _directory_resolve_content_path(request, sha1_git): """ Internal endpoint redirecting to data url for a specific file path relative to a root directory. """ try: path = os.path.normpath(request.GET.get("path")) if not path.startswith("../"): dir_info = archive.lookup_directory_with_path(sha1_git, path) if dir_info["type"] == "file": sha1 = dir_info["checksums"]["sha1"] data_url = reverse( "browse-content-raw", url_args={"query_string": sha1} ) return redirect(data_url) except Exception as exc: sentry_sdk.capture_exception(exc) return HttpResponse(status=404) diff --git a/swh/web/browse/views/release.py b/swh/web/browse/views/release.py index 9ff72c83..99f8a765 100644 --- a/swh/web/browse/views/release.py +++ b/swh/web/browse/views/release.py @@ -1,245 +1,245 @@ # Copyright (C) 2017-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import sentry_sdk from django.shortcuts import render -from swh.model.identifiers import ObjectType +from swh.model.swhids import ObjectType from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import ( gen_content_link, gen_directory_link, gen_link, gen_person_mail_link, gen_release_link, gen_revision_link, ) from swh.web.common import archive from swh.web.common.exc import NotFoundExc from swh.web.common.identifiers import get_swhids_info from swh.web.common.typing import ReleaseMetadata, SWHObjectInfo from swh.web.common.utils import format_utc_iso_date, reverse @browse_route( r"release/(?P[0-9a-f]+)/", view_name="browse-release", checksum_args=["sha1_git"], ) def release_browse(request, sha1_git): """ Django view that produces an HTML display of a release identified by its id. The url that points to it is :http:get:`/browse/release/(sha1_git)/`. """ release = archive.lookup_release(sha1_git) snapshot_context = {} origin_info = None snapshot_id = request.GET.get("snapshot_id") if not snapshot_id: snapshot_id = request.GET.get("snapshot") origin_url = request.GET.get("origin_url") if not origin_url: origin_url = request.GET.get("origin") timestamp = request.GET.get("timestamp") visit_id = request.GET.get("visit_id") if origin_url: try: snapshot_context = get_snapshot_context( snapshot_id, origin_url, timestamp, visit_id, release_name=release["name"], ) except NotFoundExc as e: raw_rel_url = reverse("browse-release", url_args={"sha1_git": sha1_git}) error_message = ( "The Software Heritage archive has a release " "with the hash you provided but the origin " "mentioned in your request appears broken: %s. " "Please check the URL and try again.\n\n" "Nevertheless, you can still browse the release " "without origin information: %s" % (gen_link(origin_url), gen_link(raw_rel_url)) ) if str(e).startswith("Origin"): raise NotFoundExc(error_message) else: raise e origin_info = snapshot_context["origin_info"] elif snapshot_id: snapshot_context = get_snapshot_context( snapshot_id, release_name=release["name"] ) snapshot_id = snapshot_context.get("snapshot_id", None) release_metadata = ReleaseMetadata( object_type=ObjectType.RELEASE, object_id=sha1_git, release=sha1_git, author=release["author"]["fullname"] if release["author"] else "None", author_url=gen_person_mail_link(release["author"]) if release["author"] else "None", date=format_utc_iso_date(release["date"]), name=release["name"], synthetic=release["synthetic"], target=release["target"], target_type=release["target_type"], snapshot=snapshot_id, origin_url=origin_url, ) release_note_lines = [] if release["message"]: release_note_lines = release["message"].split("\n") swh_objects = [SWHObjectInfo(object_type=ObjectType.RELEASE, object_id=sha1_git)] vault_cooking = None rev_directory = None target_link = None if release["target_type"] == ObjectType.REVISION.name.lower(): target_link = gen_revision_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) try: revision = archive.lookup_revision(release["target"]) rev_directory = revision["directory"] vault_cooking = { "directory_context": True, "directory_swhid": f"swh:1:dir:{rev_directory}", "revision_context": True, "revision_swhid": f"swh:1:rev:{release['target']}", } swh_objects.append( SWHObjectInfo( object_type=ObjectType.REVISION, object_id=release["target"] ) ) swh_objects.append( SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=rev_directory) ) except Exception as exc: sentry_sdk.capture_exception(exc) elif release["target_type"] == ObjectType.DIRECTORY.name.lower(): target_link = gen_directory_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) try: # check directory exists archive.lookup_directory(release["target"]) vault_cooking = { "directory_context": True, "directory_swhid": f"swh:1:dir:{release['target']}", "revision_context": False, "revision_swhid": None, } swh_objects.append( SWHObjectInfo( object_type=ObjectType.DIRECTORY, object_id=release["target"] ) ) except Exception as exc: sentry_sdk.capture_exception(exc) elif release["target_type"] == ObjectType.CONTENT.name.lower(): target_link = gen_content_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) swh_objects.append( SWHObjectInfo(object_type=ObjectType.CONTENT, object_id=release["target"]) ) elif release["target_type"] == ObjectType.RELEASE.name.lower(): target_link = gen_release_link( release["target"], snapshot_context=snapshot_context, link_text=None, link_attrs=None, ) rev_directory_url = None if rev_directory is not None: if origin_info: rev_directory_url = reverse( "browse-origin-directory", query_params={ "origin_url": origin_info["url"], "release": release["name"], "snapshot": snapshot_id, }, ) elif snapshot_id: rev_directory_url = reverse( "browse-snapshot-directory", url_args={"snapshot_id": snapshot_id}, query_params={"release": release["name"]}, ) else: rev_directory_url = reverse( "browse-directory", url_args={"sha1_git": rev_directory} ) directory_link = None if rev_directory_url is not None: directory_link = gen_link(rev_directory_url, rev_directory) release["directory_link"] = directory_link release["target_link"] = target_link if snapshot_context: snapshot_id = snapshot_context["snapshot_id"] if snapshot_id: swh_objects.append( SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id) ) swhids_info = get_swhids_info(swh_objects, snapshot_context) note_header = "None" if len(release_note_lines) > 0: note_header = release_note_lines[0] release["note_header"] = note_header release["note_body"] = "\n".join(release_note_lines[1:]) heading = "Release - %s" % release["name"] if snapshot_context: context_found = "snapshot: %s" % snapshot_context["snapshot_id"] if origin_info: context_found = "origin: %s" % origin_info["url"] heading += " - %s" % context_found return render( request, "browse/release.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"], "swh_object_name": "Release", "swh_object_metadata": release_metadata, "release": release, "snapshot_context": snapshot_context, "show_actions": True, "breadcrumbs": None, "vault_cooking": vault_cooking, "top_right_link": None, "swhids_info": swhids_info, }, ) diff --git a/swh/web/browse/views/revision.py b/swh/web/browse/views/revision.py index dd5eb303..66ace89f 100644 --- a/swh/web/browse/views/revision.py +++ b/swh/web/browse/views/revision.py @@ -1,586 +1,586 @@ # Copyright (C) 2017-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import hashlib import json import textwrap from django.http import JsonResponse from django.shortcuts import render from django.utils.safestring import mark_safe from swh.model.hashutil import hash_to_bytes -from swh.model.identifiers import CoreSWHID, ObjectType +from swh.model.swhids import CoreSWHID, ObjectType from swh.web.browse.browseurls import browse_route from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import ( content_display_max_size, format_log_entries, gen_link, gen_person_mail_link, gen_revision_url, get_directory_entries, get_readme_to_display, get_revision_log_url, prepare_content_for_display, request_content, ) from swh.web.common import archive from swh.web.common.exc import NotFoundExc, http_status_code_message from swh.web.common.identifiers import get_swhids_info from swh.web.common.typing import RevisionMetadata, SWHObjectInfo from swh.web.common.utils import ( format_utc_iso_date, gen_path_info, reverse, swh_object_icons, ) def _gen_content_url(revision, query_string, path, snapshot_context): if snapshot_context: query_params = snapshot_context["query_params"] query_params["path"] = path query_params["revision"] = revision["id"] content_url = reverse("browse-origin-content", query_params=query_params) else: content_path = "%s/%s" % (revision["directory"], path) content_url = reverse( "browse-content", url_args={"query_string": query_string}, query_params={"path": content_path}, ) return content_url def _gen_diff_link(idx, diff_anchor, link_text): if idx < _max_displayed_file_diffs: return gen_link(diff_anchor, link_text) else: return link_text # TODO: put in conf _max_displayed_file_diffs = 1000 def _gen_revision_changes_list(revision, changes, snapshot_context): """ Returns a HTML string describing the file changes introduced in a revision. As this string will be displayed in the browse revision view, links to adequate file diffs are also generated. Args: revision (str): hexadecimal representation of a revision identifier changes (list): list of file changes in the revision snapshot_context (dict): optional origin context used to reverse the content urls Returns: A string to insert in a revision HTML view. """ changes_msg = [] for i, change in enumerate(changes): hasher = hashlib.sha1() from_query_string = "" to_query_string = "" diff_id = "diff-" if change["from"]: from_query_string = "sha1_git:" + change["from"]["target"] diff_id += change["from"]["target"] + "-" + change["from_path"] diff_id += "-" if change["to"]: to_query_string = "sha1_git:" + change["to"]["target"] diff_id += change["to"]["target"] + change["to_path"] change["path"] = change["to_path"] or change["from_path"] url_args = { "from_query_string": from_query_string, "to_query_string": to_query_string, } query_params = {"path": change["path"]} change["diff_url"] = reverse( "diff-contents", url_args=url_args, query_params=query_params ) hasher.update(diff_id.encode("utf-8")) diff_id = hasher.hexdigest() change["id"] = diff_id diff_link = "#diff_" + diff_id if change["type"] == "modify": change["content_url"] = _gen_content_url( revision, to_query_string, change["to_path"], snapshot_context ) changes_msg.append( "modified: %s" % _gen_diff_link(i, diff_link, change["to_path"]) ) elif change["type"] == "insert": change["content_url"] = _gen_content_url( revision, to_query_string, change["to_path"], snapshot_context ) changes_msg.append( "new file: %s" % _gen_diff_link(i, diff_link, change["to_path"]) ) elif change["type"] == "delete": parent = archive.lookup_revision(revision["parents"][0]) change["content_url"] = _gen_content_url( parent, from_query_string, change["from_path"], snapshot_context ) changes_msg.append( "deleted: %s" % _gen_diff_link(i, diff_link, change["from_path"]) ) elif change["type"] == "rename": change["content_url"] = _gen_content_url( revision, to_query_string, change["to_path"], snapshot_context ) link_text = change["from_path"] + " → " + change["to_path"] changes_msg.append( "renamed: %s" % _gen_diff_link(i, diff_link, link_text) ) if not changes: changes_msg.append("No changes") return mark_safe("\n".join(changes_msg)) @browse_route( r"revision/(?P[0-9a-f]+)/diff/", view_name="diff-revision", checksum_args=["sha1_git"], ) def _revision_diff(request, sha1_git): """ Browse internal endpoint to compute revision diff """ revision = archive.lookup_revision(sha1_git) snapshot_context = None origin_url = request.GET.get("origin_url", None) if not origin_url: origin_url = request.GET.get("origin", None) timestamp = request.GET.get("timestamp", None) visit_id = request.GET.get("visit_id", None) if origin_url: snapshot_context = get_snapshot_context( origin_url=origin_url, timestamp=timestamp, visit_id=visit_id ) changes = archive.diff_revision(sha1_git) changes_msg = _gen_revision_changes_list(revision, changes, snapshot_context) diff_data = { "total_nb_changes": len(changes), "changes": changes[:_max_displayed_file_diffs], "changes_msg": changes_msg, } return JsonResponse(diff_data) NB_LOG_ENTRIES = 100 @browse_route( r"revision/(?P[0-9a-f]+)/log/", view_name="browse-revision-log", checksum_args=["sha1_git"], ) def revision_log_browse(request, sha1_git): """ Django view that produces an HTML display of the history log for a revision identified by its id. The url that points to it is :http:get:`/browse/revision/(sha1_git)/log/` """ origin_url = request.GET.get("origin_url") snapshot_id = request.GET.get("snapshot") snapshot_context = None if origin_url or snapshot_id: snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=request.GET.get("timestamp"), visit_id=request.GET.get("visit_id"), branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=sha1_git, ) per_page = int(request.GET.get("per_page", NB_LOG_ENTRIES)) offset = int(request.GET.get("offset", 0)) revs_ordering = request.GET.get("revs_ordering", "committer_date") session_key = "rev_%s_log_ordering_%s" % (sha1_git, revs_ordering) rev_log_session = request.session.get(session_key, None) rev_log = [] revs_walker_state = None if rev_log_session: rev_log = rev_log_session["rev_log"] revs_walker_state = rev_log_session["revs_walker_state"] if len(rev_log) < offset + per_page: revs_walker = archive.get_revisions_walker( revs_ordering, sha1_git, max_revs=offset + per_page + 1, state=revs_walker_state, ) rev_log += [rev["id"] for rev in revs_walker] revs_walker_state = revs_walker.export_state() revs = rev_log[offset : offset + per_page] revision_log = archive.lookup_revision_multiple(revs) request.session[session_key] = { "rev_log": rev_log, "revs_walker_state": revs_walker_state, } revs_ordering = request.GET.get("revs_ordering", "") prev_log_url = None if len(rev_log) > offset + per_page: prev_log_url = reverse( "browse-revision-log", url_args={"sha1_git": sha1_git}, query_params={ "per_page": per_page, "offset": offset + per_page, "revs_ordering": revs_ordering or None, }, ) next_log_url = None if offset != 0: next_log_url = reverse( "browse-revision-log", url_args={"sha1_git": sha1_git}, query_params={ "per_page": per_page, "offset": offset - per_page, "revs_ordering": revs_ordering or None, }, ) revision_log_data = format_log_entries(revision_log, per_page) swh_rev_id = str( CoreSWHID(object_type=ObjectType.REVISION, object_id=hash_to_bytes(sha1_git)) ) return render( request, "browse/revision-log.html", { "heading": "Revision history", "swh_object_id": swh_rev_id, "swh_object_name": "Revisions history", "swh_object_metadata": None, "revision_log": revision_log_data, "revs_ordering": revs_ordering, "next_log_url": next_log_url, "prev_log_url": prev_log_url, "breadcrumbs": None, "top_right_link": None, "snapshot_context": snapshot_context, "vault_cooking": None, "show_actions": True, "swhids_info": None, }, ) @browse_route( r"revision/(?P[0-9a-f]+)/", view_name="browse-revision", checksum_args=["sha1_git"], ) def revision_browse(request, sha1_git): """ Django view that produces an HTML display of a revision identified by its id. The url that points to it is :http:get:`/browse/revision/(sha1_git)/`. """ revision = archive.lookup_revision(sha1_git) origin_info = None snapshot_context = None origin_url = request.GET.get("origin_url") if not origin_url: origin_url = request.GET.get("origin") timestamp = request.GET.get("timestamp") visit_id = request.GET.get("visit_id") snapshot_id = request.GET.get("snapshot_id") if not snapshot_id: snapshot_id = request.GET.get("snapshot") path = request.GET.get("path") dir_id = None dirs, files = [], [] content_data = {} if origin_url: try: snapshot_context = get_snapshot_context( snapshot_id=snapshot_id, origin_url=origin_url, timestamp=timestamp, visit_id=visit_id, branch_name=request.GET.get("branch"), release_name=request.GET.get("release"), revision_id=sha1_git, path=path, ) except NotFoundExc as e: raw_rev_url = reverse("browse-revision", url_args={"sha1_git": sha1_git}) error_message = ( "The Software Heritage archive has a revision " "with the hash you provided but the origin " "mentioned in your request appears broken: %s. " "Please check the URL and try again.\n\n" "Nevertheless, you can still browse the revision " "without origin information: %s" % (gen_link(origin_url), gen_link(raw_rev_url)) ) if str(e).startswith("Origin"): raise NotFoundExc(error_message) else: raise e origin_info = snapshot_context["origin_info"] snapshot_id = snapshot_context["snapshot_id"] elif snapshot_id: snapshot_context = get_snapshot_context(snapshot_id) error_info = {"status_code": 200, "description": None} if path: try: file_info = archive.lookup_directory_with_path(revision["directory"], path) if file_info["type"] == "dir": dir_id = file_info["target"] else: query_string = "sha1_git:" + file_info["target"] content_data = request_content(query_string) except NotFoundExc as e: error_info["status_code"] = 404 error_info["description"] = f"NotFoundExc: {str(e)}" else: dir_id = revision["directory"] if dir_id: path = "" if path is None else (path + "/") dirs, files = get_directory_entries(dir_id) revision_metadata = RevisionMetadata( object_type=ObjectType.REVISION, object_id=sha1_git, revision=sha1_git, author=revision["author"]["fullname"] if revision["author"] else "None", author_url=gen_person_mail_link(revision["author"]) if revision["author"] else "None", committer=revision["committer"]["fullname"] if revision["committer"] else "None", committer_url=gen_person_mail_link(revision["committer"]) if revision["committer"] else "None", committer_date=format_utc_iso_date(revision["committer_date"]), date=format_utc_iso_date(revision["date"]), directory=revision["directory"], merge=revision["merge"], metadata=json.dumps( revision["metadata"], sort_keys=True, indent=4, separators=(",", ": ") ), parents=revision["parents"], synthetic=revision["synthetic"], type=revision["type"], snapshot=snapshot_id, origin_url=origin_url, ) message_lines = ["None"] if revision["message"]: message_lines = revision["message"].split("\n") parents = [] for p in revision["parents"]: parent_url = gen_revision_url(p, snapshot_context) parents.append({"id": p, "url": parent_url}) path_info = gen_path_info(path) query_params = snapshot_context["query_params"] if snapshot_context else {} breadcrumbs = [] breadcrumbs.append( { "name": revision["directory"][:7], "url": reverse( "browse-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ), } ) for pi in path_info: query_params["path"] = pi["path"] breadcrumbs.append( { "name": pi["name"], "url": reverse( "browse-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ), } ) vault_cooking = { "directory_context": False, "directory_swhid": None, "revision_context": True, "revision_swhid": f"swh:1:rev:{sha1_git}", } swh_objects = [SWHObjectInfo(object_type=ObjectType.REVISION, object_id=sha1_git)] content = None content_size = None filename = None mimetype = None language = None readme_name = None readme_url = None readme_html = None readmes = {} extra_context = dict(revision_metadata) extra_context["path"] = f"/{path}" if path else None if content_data: breadcrumbs[-1]["url"] = None content_size = content_data["length"] mimetype = content_data["mimetype"] if content_data["raw_data"]: content_display_data = prepare_content_for_display( content_data["raw_data"], content_data["mimetype"], path ) content = content_display_data["content_data"] language = content_display_data["language"] mimetype = content_display_data["mimetype"] if path: filename = path_info[-1]["name"] query_params["filename"] = filename filepath = "/".join(pi["name"] for pi in path_info[:-1]) extra_context["path"] = f"/{filepath}/" if filepath else "/" extra_context["filename"] = filename top_right_link = { "url": reverse( "browse-content-raw", url_args={"query_string": query_string}, query_params={"filename": filename}, ), "icon": swh_object_icons["content"], "text": "Raw File", } swh_objects.append( SWHObjectInfo(object_type=ObjectType.CONTENT, object_id=file_info["target"]) ) else: for d in dirs: if d["type"] == "rev": d["url"] = reverse( "browse-revision", url_args={"sha1_git": d["target"]} ) else: query_params["path"] = path + d["name"] d["url"] = reverse( "browse-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ) for f in files: query_params["path"] = path + f["name"] f["url"] = reverse( "browse-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ) if f["name"].lower().startswith("readme"): readmes[f["name"]] = f["checksums"]["sha1"] readme_name, readme_url, readme_html = get_readme_to_display(readmes) top_right_link = { "url": get_revision_log_url(sha1_git, snapshot_context), "icon": swh_object_icons["revisions history"], "text": "History", } vault_cooking["directory_context"] = True vault_cooking["directory_swhid"] = f"swh:1:dir:{dir_id}" swh_objects.append( SWHObjectInfo(object_type=ObjectType.DIRECTORY, object_id=dir_id) ) query_params.pop("path", None) diff_revision_url = reverse( "diff-revision", url_args={"sha1_git": sha1_git}, query_params=query_params, ) if snapshot_id: swh_objects.append( SWHObjectInfo(object_type=ObjectType.SNAPSHOT, object_id=snapshot_id) ) swhids_info = get_swhids_info(swh_objects, snapshot_context, extra_context) heading = "Revision - %s - %s" % ( sha1_git[:7], textwrap.shorten(message_lines[0], width=70), ) if snapshot_context: context_found = "snapshot: %s" % snapshot_context["snapshot_id"] if origin_info: context_found = "origin: %s" % origin_info["url"] heading += " - %s" % context_found return render( request, "browse/revision.html", { "heading": heading, "swh_object_id": swhids_info[0]["swhid"], "swh_object_name": "Revision", "swh_object_metadata": revision_metadata, "message_header": message_lines[0], "message_body": "\n".join(message_lines[1:]), "parents": parents, "snapshot_context": snapshot_context, "dirs": dirs, "files": files, "content": content, "content_size": content_size, "max_content_size": content_display_max_size, "filename": filename, "encoding": content_data.get("encoding"), "mimetype": mimetype, "language": language, "readme_name": readme_name, "readme_url": readme_url, "readme_html": readme_html, "breadcrumbs": breadcrumbs, "top_right_link": top_right_link, "vault_cooking": vault_cooking, "diff_revision_url": diff_revision_url, "show_actions": True, "swhids_info": swhids_info, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], }, status=error_info["status_code"], ) diff --git a/swh/web/common/archive.py b/swh/web/common/archive.py index 17a6ad8d..8ac97e0f 100644 --- a/swh/web/common/archive.py +++ b/swh/web/common/archive.py @@ -1,1405 +1,1405 @@ # Copyright (C) 2015-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from collections import defaultdict import itertools import os import re from typing import Any, Dict, Iterable, Iterator, List, Optional, Set, Tuple, Union from urllib.parse import urlparse from swh.model import hashutil -from swh.model.identifiers import CoreSWHID, ObjectType from swh.model.model import OriginVisit, Revision +from swh.model.swhids import CoreSWHID, ObjectType from swh.storage.algos import diff, revisions_walker from swh.storage.algos.origin import origin_get_latest_visit_status from swh.storage.algos.snapshot import snapshot_get_latest, snapshot_resolve_alias from swh.vault.exc import NotFoundExc as VaultNotFoundExc from swh.web import config from swh.web.common import converters, query from swh.web.common.exc import NotFoundExc from swh.web.common.typing import ( OriginInfo, OriginMetadataInfo, OriginVisitInfo, PagedResult, ) search = config.search() storage = config.storage() vault = config.vault() idx_storage = config.indexer_storage() counters = config.counters() MAX_LIMIT = 50 # Top limit the users can ask for def _first_element(lst): """Returns the first element in the provided list or None if it is empty or None""" return next(iter(lst or []), None) def lookup_multiple_hashes(hashes): """Lookup the passed hashes in a single DB connection, using batch processing. Args: An array of {filename: X, sha1: Y}, string X, hex sha1 string Y. Returns: The same array with elements updated with elem['found'] = true if the hash is present in storage, elem['found'] = false if not. """ hashlist = [hashutil.hash_to_bytes(elem["sha1"]) for elem in hashes] content_missing = storage.content_missing_per_sha1(hashlist) missing = [hashutil.hash_to_hex(x) for x in content_missing] for x in hashes: x.update({"found": True}) for h in hashes: if h["sha1"] in missing: h["found"] = False return hashes def lookup_expression(expression, last_sha1, per_page): """Lookup expression in raw content. Args: expression (str): An expression to lookup through raw indexed content last_sha1 (str): Last sha1 seen per_page (int): Number of results per page Yields: ctags whose content match the expression """ limit = min(per_page, MAX_LIMIT) ctags = idx_storage.content_ctags_search( expression, last_sha1=last_sha1, limit=limit ) for ctag in ctags: ctag = converters.from_swh(ctag, hashess={"id"}) ctag["sha1"] = ctag["id"] ctag.pop("id") yield ctag def lookup_hash(q: str) -> Dict[str, Any]: """Check if the storage contains a given content checksum and return it if found. Args: q: query string of the form Returns: Dict with key found containing the hash info if the hash is present, None if not. """ algo, hash_ = query.parse_hash(q) found = _first_element(storage.content_find({algo: hash_})) if found: content = converters.from_content(found.to_dict()) else: content = None return {"found": content, "algo": algo} def search_hash(q: str) -> Dict[str, bool]: """Search storage for a given content checksum. Args: q: query string of the form Returns: Dict with key found to True or False, according to whether the checksum is present or not """ algo, hash_ = query.parse_hash(q) found = _first_element(storage.content_find({algo: hash_})) return {"found": found is not None} def _lookup_content_sha1(q: str) -> Optional[bytes]: """Given a possible input, query for the content's sha1. Args: q: query string of the form Returns: binary sha1 if found or None """ algo, hash_ = query.parse_hash(q) if algo != "sha1": hashes = _first_element(storage.content_find({algo: hash_})) if not hashes: return None return hashes.sha1 return hash_ def lookup_content_ctags(q): """Return ctags information from a specified content. Args: q: query string of the form Yields: ctags information (dict) list if the content is found. """ sha1 = _lookup_content_sha1(q) if not sha1: return None ctags = list(idx_storage.content_ctags_get([sha1])) if not ctags: return None for ctag in ctags: yield converters.from_swh(ctag, hashess={"id"}) def lookup_content_filetype(q): """Return filetype information from a specified content. Args: q: query string of the form Yields: filetype information (dict) list if the content is found. """ sha1 = _lookup_content_sha1(q) if not sha1: return None filetype = _first_element(list(idx_storage.content_mimetype_get([sha1]))) if not filetype: return None return converters.from_filetype(filetype.to_dict()) def lookup_content_language(q): """Always returns None. This used to return language information from a specified content, but this is currently disabled. Args: q: query string of the form Yields: language information (dict) list if the content is found. """ return None def lookup_content_license(q): """Return license information from a specified content. Args: q: query string of the form Yields: license information (dict) list if the content is found. """ sha1 = _lookup_content_sha1(q) if not sha1: return None licenses = list(idx_storage.content_fossology_license_get([sha1])) if not licenses: return None license_dicts = [license.to_dict() for license in licenses] for license_dict in license_dicts: del license_dict["id"] lic = { "id": sha1, "facts": license_dicts, } return converters.from_swh(lic, hashess={"id"}) def lookup_origin(origin: OriginInfo) -> OriginInfo: """Return information about the origin matching dict origin. Args: origin: origin's dict with 'url' key Returns: origin information as dict. """ origin_urls = [origin["url"]] if origin["url"]: # handle case when user provided an origin url with a trailing # slash while the url in storage does not have it (e.g. GitHub) if origin["url"].endswith("/"): origin_urls.append(origin["url"][:-1]) # handle case when user provided an origin url without a trailing # slash while the url in storage have it (e.g. Debian source package) else: origin_urls.append(f"{origin['url']}/") try: # handle case where the "://" character sequence was mangled into ":/" parsed_url = urlparse(origin["url"]) if ( parsed_url.scheme and not parsed_url.netloc and origin["url"].startswith(f"{parsed_url.scheme}:/") and not origin["url"].startswith(f"{parsed_url.scheme}://") ): origin_urls.append( origin["url"].replace( f"{parsed_url.scheme}:/", f"{parsed_url.scheme}://" ) ) except Exception: pass origins = [o for o in storage.origin_get(origin_urls) if o is not None] if not origins: msg = "Origin with url %s not found!" % origin["url"] raise NotFoundExc(msg) return converters.from_origin(origins[0].to_dict()) def lookup_origins( page_token: Optional[str], limit: int = 100 ) -> PagedResult[OriginInfo]: """Get list of archived software origins in a paginated way. Origins are sorted by id before returning them Args: origin_from (int): The minimum id of the origins to return origin_count (int): The maximum number of origins to return Returns: Page of OriginInfo """ page = storage.origin_list(page_token=page_token, limit=limit) return PagedResult( [converters.from_origin(o.to_dict()) for o in page.results], next_page_token=page.next_page_token, ) def search_origin( url_pattern: str, use_ql: bool = False, limit: int = 50, with_visit: bool = False, visit_types: Optional[List[str]] = None, page_token: Optional[str] = None, ) -> Tuple[List[OriginInfo], Optional[str]]: """Search for origins whose urls contain a provided string pattern or match a provided regular expression. Args: url_pattern: the string pattern to search for in origin urls use_ql: whether to use swh search query language or not limit: the maximum number of found origins to return with_visit: Whether origins with no visit are to be filtered out visit_types: Only origins having any of the provided visit types (e.g. git, svn, pypi) will be returned page_token: opaque string used to get the next results of a search Returns: list of origin information as dict. """ if page_token: assert isinstance(page_token, str) if search: if use_ql: page_result = search.origin_search( query=url_pattern, page_token=page_token, with_visit=with_visit, visit_types=visit_types, limit=limit, ) else: page_result = search.origin_search( url_pattern=url_pattern, page_token=page_token, with_visit=with_visit, visit_types=visit_types, limit=limit, ) origins = [converters.from_origin(ori_dict) for ori_dict in page_result.results] else: # Fallback to swh-storage if swh-search is not configured search_words = [re.escape(word) for word in url_pattern.split()] if len(search_words) >= 7: url_pattern = ".*".join(search_words) else: pattern_parts = [] for permut in itertools.permutations(search_words): pattern_parts.append(".*".join(permut)) url_pattern = "|".join(pattern_parts) page_result = storage.origin_search( url_pattern, page_token=page_token, with_visit=with_visit, limit=limit, visit_types=visit_types, regexp=True, ) origins = [converters.from_origin(ori.to_dict()) for ori in page_result.results] return (origins, page_result.next_page_token) def search_origin_metadata( fulltext: str, limit: int = 50 ) -> Iterable[OriginMetadataInfo]: """Search for origins whose metadata match a provided string pattern. Args: fulltext: the string pattern to search for in origin metadata limit: the maximum number of found origins to return Returns: Iterable of origin metadata information for existing origins """ results = [] if ( search and config.get_config()["search_config"]["metadata_backend"] == "swh-search" ): page_result = search.origin_search(metadata_pattern=fulltext, limit=limit,) matches = idx_storage.origin_intrinsic_metadata_get( [r["url"] for r in page_result.results] ) else: matches = idx_storage.origin_intrinsic_metadata_search_fulltext( conjunction=[fulltext], limit=limit ) matches = [match.to_dict() for match in matches] origins = storage.origin_get([match["id"] for match in matches]) for origin, match in zip(origins, matches): if not origin: continue match["from_revision"] = hashutil.hash_to_hex(match["from_revision"]) del match["id"] results.append(OriginMetadataInfo(url=origin.url, metadata=match)) return results def lookup_origin_intrinsic_metadata(origin_url: str) -> Dict[str, Any]: """Return intrinsic metadata for origin whose origin matches given origin. Args: origin_url: origin url Raises: NotFoundExc when the origin is not found Returns: origin metadata. """ origins = [origin_url] origin_info = storage.origin_get(origins)[0] if not origin_info: raise NotFoundExc(f"Origin with url {origin_url} not found!") match = _first_element(idx_storage.origin_intrinsic_metadata_get(origins)) result = {} if match: result = match.metadata return result def _to_sha1_bin(sha1_hex): _, sha1_git_bin = query.parse_hash_with_algorithms_or_throws( sha1_hex, ["sha1"], "Only sha1_git is supported." # HACK: sha1_git really ) return sha1_git_bin def _check_directory_exists(sha1_git, sha1_git_bin): if len(list(storage.directory_missing([sha1_git_bin]))): raise NotFoundExc("Directory with sha1_git %s not found" % sha1_git) def lookup_directory(sha1_git): """Return information about the directory with id sha1_git. Args: sha1_git as string Returns: directory information as dict. """ empty_dir_sha1 = "4b825dc642cb6eb9a060e54bf8d69288fbee4904" if sha1_git == empty_dir_sha1: return [] sha1_git_bin = _to_sha1_bin(sha1_git) _check_directory_exists(sha1_git, sha1_git_bin) directory_entries = storage.directory_ls(sha1_git_bin) return map(converters.from_directory_entry, directory_entries) def lookup_directory_with_path(sha1_git: str, path: str) -> Dict[str, Any]: """Return directory information for entry with specified path w.r.t. root directory pointed by sha1_git Args: sha1_git: sha1_git corresponding to the directory to which we append paths to (hopefully) find the entry path: the relative path to the entry starting from the root directory pointed by sha1_git Returns: Directory entry information as dict. Raises: NotFoundExc if the directory entry is not found """ sha1_git_bin = _to_sha1_bin(sha1_git) _check_directory_exists(sha1_git, sha1_git_bin) paths = path.strip(os.path.sep).split(os.path.sep) queried_dir = storage.directory_entry_get_by_path( sha1_git_bin, [p.encode("utf-8") for p in paths] ) if not queried_dir: raise NotFoundExc( f"Directory entry with path {path} from root directory {sha1_git} not found" ) return converters.from_directory_entry(queried_dir) def lookup_release(release_sha1_git: str) -> Dict[str, Any]: """Return information about the release with sha1 release_sha1_git. Args: release_sha1_git: The release's sha1 as hexadecimal Returns: Release information as dict. Raises: ValueError if the identifier provided is not of sha1 nature. """ sha1_git_bin = _to_sha1_bin(release_sha1_git) release = _first_element(storage.release_get([sha1_git_bin])) if not release: raise NotFoundExc(f"Release with sha1_git {release_sha1_git} not found.") return converters.from_release(release) def lookup_release_multiple(sha1_git_list) -> Iterator[Optional[Dict[str, Any]]]: """Return information about the releases identified with their sha1_git identifiers. Args: sha1_git_list: A list of release sha1_git identifiers Returns: Iterator of Release metadata information as dict. Raises: ValueError if the identifier provided is not of sha1 nature. """ sha1_bin_list = [_to_sha1_bin(sha1_git) for sha1_git in sha1_git_list] releases = storage.release_get(sha1_bin_list) for r in releases: if r is not None: yield converters.from_release(r) else: yield None def lookup_revision(rev_sha1_git) -> Dict[str, Any]: """Return information about the revision with sha1 revision_sha1_git. Args: revision_sha1_git: The revision's sha1 as hexadecimal Returns: Revision information as dict. Raises: ValueError if the identifier provided is not of sha1 nature. NotFoundExc if there is no revision with the provided sha1_git. """ sha1_git_bin = _to_sha1_bin(rev_sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision with sha1_git {rev_sha1_git} not found.") return converters.from_revision(revision) def lookup_revision_multiple(sha1_git_list) -> Iterator[Optional[Dict[str, Any]]]: """Return information about the revisions identified with their sha1_git identifiers. Args: sha1_git_list: A list of revision sha1_git identifiers Yields: revision information as dict if the revision exists, None otherwise. Raises: ValueError if the identifier provided is not of sha1 nature. """ sha1_bin_list = [_to_sha1_bin(sha1_git) for sha1_git in sha1_git_list] revisions = storage.revision_get(sha1_bin_list) for revision in revisions: if revision is not None: yield converters.from_revision(revision) else: yield None def lookup_revision_message(rev_sha1_git) -> Dict[str, bytes]: """Return the raw message of the revision with sha1 revision_sha1_git. Args: revision_sha1_git: The revision's sha1 as hexadecimal Returns: Decoded revision message as dict {'message': } Raises: ValueError if the identifier provided is not of sha1 nature. NotFoundExc if the revision is not found, or if it has no message """ sha1_git_bin = _to_sha1_bin(rev_sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision with sha1_git {rev_sha1_git} not found.") if not revision.message: raise NotFoundExc(f"No message for revision with sha1_git {rev_sha1_git}.") return {"message": revision.message} def _lookup_revision_id_by(origin, branch_name, timestamp): def _get_snapshot_branch(snapshot, branch_name): snapshot = lookup_snapshot( visit["snapshot"], branches_from=branch_name, branches_count=10, branch_name_exclude_prefix=None, ) branch = None if branch_name in snapshot["branches"]: branch = snapshot["branches"][branch_name] return branch if isinstance(origin, int): origin = {"id": origin} elif isinstance(origin, str): origin = {"url": origin} else: raise TypeError('"origin" must be an int or a string.') from swh.web.common.origin_visits import get_origin_visit visit = get_origin_visit(origin, visit_ts=timestamp) branch = _get_snapshot_branch(visit["snapshot"], branch_name) rev_id = None if branch and branch["target_type"] == "revision": rev_id = branch["target"] elif branch and branch["target_type"] == "alias": branch = _get_snapshot_branch(visit["snapshot"], branch["target"]) if branch and branch["target_type"] == "revision": rev_id = branch["target"] if not rev_id: raise NotFoundExc( "Revision for origin %s and branch %s not found." % (origin.get("url"), branch_name) ) return rev_id def lookup_revision_by(origin, branch_name="HEAD", timestamp=None): """Lookup revision by origin, snapshot branch name and visit timestamp. If branch_name is not provided, lookup using 'HEAD' as default. If timestamp is not provided, use the most recent. Args: origin (Union[int,str]): origin of the revision branch_name (str): snapshot branch name timestamp (str/int): origin visit time frame Returns: dict: The revision matching the criterions Raises: NotFoundExc if no revision corresponds to the criterion """ rev_id = _lookup_revision_id_by(origin, branch_name, timestamp) return lookup_revision(rev_id) def lookup_revision_log(rev_sha1_git, limit): """Lookup revision log by revision id. Args: rev_sha1_git (str): The revision's sha1 as hexadecimal limit (int): the maximum number of revisions returned Returns: list: Revision log as list of revision dicts Raises: ValueError: if the identifier provided is not of sha1 nature. swh.web.common.exc.NotFoundExc: if there is no revision with the provided sha1_git. """ lookup_revision(rev_sha1_git) sha1_git_bin = _to_sha1_bin(rev_sha1_git) revision_entries = storage.revision_log([sha1_git_bin], limit) return map(converters.from_revision, revision_entries) def lookup_revision_log_by(origin, branch_name, timestamp, limit): """Lookup revision by origin, snapshot branch name and visit timestamp. Args: origin (Union[int,str]): origin of the revision branch_name (str): snapshot branch timestamp (str/int): origin visit time frame limit (int): the maximum number of revisions returned Returns: list: Revision log as list of revision dicts Raises: swh.web.common.exc.NotFoundExc: if no revision corresponds to the criterion """ rev_id = _lookup_revision_id_by(origin, branch_name, timestamp) return lookup_revision_log(rev_id, limit) def lookup_revision_with_context_by( origin, branch_name, timestamp, sha1_git, limit=100 ): """Return information about revision sha1_git, limited to the sub-graph of all transitive parents of sha1_git_root. sha1_git_root being resolved through the lookup of a revision by origin, branch_name and ts. In other words, sha1_git is an ancestor of sha1_git_root. Args: - origin: origin of the revision. - branch_name: revision's branch. - timestamp: revision's time frame. - sha1_git: one of sha1_git_root's ancestors. - limit: limit the lookup to 100 revisions back. Returns: Pair of (root_revision, revision). Information on sha1_git if it is an ancestor of sha1_git_root including children leading to sha1_git_root Raises: - BadInputExc in case of unknown algo_hash or bad hash. - NotFoundExc if either revision is not found or if sha1_git is not an ancestor of sha1_git_root. """ rev_root_id = _lookup_revision_id_by(origin, branch_name, timestamp) rev_root_id_bin = hashutil.hash_to_bytes(rev_root_id) rev_root = storage.revision_get([rev_root_id_bin])[0] return ( converters.from_revision(rev_root) if rev_root else None, lookup_revision_with_context(rev_root, sha1_git, limit), ) def lookup_revision_with_context( sha1_git_root: Union[str, Dict[str, Any], Revision], sha1_git: str, limit: int = 100 ) -> Dict[str, Any]: """Return information about revision sha1_git, limited to the sub-graph of all transitive parents of sha1_git_root. In other words, sha1_git is an ancestor of sha1_git_root. Args: sha1_git_root: latest revision. The type is either a sha1 (as an hex string) or a non converted dict. sha1_git: one of sha1_git_root's ancestors limit: limit the lookup to 100 revisions back Returns: Information on sha1_git if it is an ancestor of sha1_git_root including children leading to sha1_git_root Raises: BadInputExc in case of unknown algo_hash or bad hash NotFoundExc if either revision is not found or if sha1_git is not an ancestor of sha1_git_root """ sha1_git_bin = _to_sha1_bin(sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision {sha1_git} not found") if isinstance(sha1_git_root, str): sha1_git_root_bin = _to_sha1_bin(sha1_git_root) revision_root = storage.revision_get([sha1_git_root_bin])[0] if not revision_root: raise NotFoundExc(f"Revision root {sha1_git_root} not found") elif isinstance(sha1_git_root, Revision): sha1_git_root_bin = sha1_git_root.id else: sha1_git_root_bin = sha1_git_root["id"] revision_log = storage.revision_log([sha1_git_root_bin], limit) parents: Dict[str, List[str]] = {} children = defaultdict(list) for rev in revision_log: rev_id = rev["id"] parents[rev_id] = [] for parent_id in rev["parents"]: parents[rev_id].append(parent_id) children[parent_id].append(rev_id) if revision.id not in parents: raise NotFoundExc(f"Revision {sha1_git} is not an ancestor of {sha1_git_root}") revision_d = revision.to_dict() revision_d["children"] = children[revision.id] return converters.from_revision(revision_d) def lookup_directory_with_revision(sha1_git, dir_path=None, with_data=False): """Return information on directory pointed by revision with sha1_git. If dir_path is not provided, display top level directory. Otherwise, display the directory pointed by dir_path (if it exists). Args: sha1_git: revision's hash. dir_path: optional directory pointed to by that revision. with_data: boolean that indicates to retrieve the raw data if the path resolves to a content. Default to False (for the api) Returns: Information on the directory pointed to by that revision. Raises: BadInputExc in case of unknown algo_hash or bad hash. NotFoundExc either if the revision is not found or the path referenced does not exist. NotImplementedError in case of dir_path exists but do not reference a type 'dir' or 'file'. """ sha1_git_bin = _to_sha1_bin(sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision {sha1_git} not found") dir_sha1_git_bin = revision.directory if dir_path: paths = dir_path.strip(os.path.sep).split(os.path.sep) entity = storage.directory_entry_get_by_path( dir_sha1_git_bin, list(map(lambda p: p.encode("utf-8"), paths)) ) if not entity: raise NotFoundExc( "Directory or File '%s' pointed to by revision %s not found" % (dir_path, sha1_git) ) else: entity = {"type": "dir", "target": dir_sha1_git_bin} if entity["type"] == "dir": directory_entries = storage.directory_ls(entity["target"]) or [] return { "type": "dir", "path": "." if not dir_path else dir_path, "revision": sha1_git, "content": list(map(converters.from_directory_entry, directory_entries)), } elif entity["type"] == "file": # content content = _first_element(storage.content_find({"sha1_git": entity["target"]})) if not content: raise NotFoundExc(f"Content not found for revision {sha1_git}") content_d = content.to_dict() if with_data: data = storage.content_get_data(content.sha1) if data: content_d["data"] = data return { "type": "file", "path": "." if not dir_path else dir_path, "revision": sha1_git, "content": converters.from_content(content_d), } elif entity["type"] == "rev": # revision revision = storage.revision_get([entity["target"]])[0] return { "type": "rev", "path": "." if not dir_path else dir_path, "revision": sha1_git, "content": converters.from_revision(revision) if revision else None, } else: raise NotImplementedError("Entity of type %s not implemented." % entity["type"]) def lookup_content(q: str) -> Dict[str, Any]: """Lookup the content designed by q. Args: q: The release's sha1 as hexadecimal Raises: NotFoundExc if the requested content is not found """ algo, hash_ = query.parse_hash(q) c = _first_element(storage.content_find({algo: hash_})) if not c: hhex = hashutil.hash_to_hex(hash_) raise NotFoundExc(f"Content with {algo} checksum equals to {hhex} not found!") return converters.from_content(c.to_dict()) def lookup_content_raw(q: str) -> Dict[str, Any]: """Lookup the content defined by q. Args: q: query string of the form Returns: dict with 'sha1' and 'data' keys. data representing its raw data decoded. Raises: NotFoundExc if the requested content is not found or if the content bytes are not available in the storage """ c = lookup_content(q) content_sha1_bytes = hashutil.hash_to_bytes(c["checksums"]["sha1"]) content_data = storage.content_get_data(content_sha1_bytes) if content_data is None: algo, hash_ = query.parse_hash(q) raise NotFoundExc( f"Bytes of content with {algo} checksum equals " f"to {hashutil.hash_to_hex(hash_)} are not available!" ) return converters.from_content({"sha1": content_sha1_bytes, "data": content_data}) def stat_counters(): """Return the stat counters for Software Heritage Returns: A dict mapping textual labels to integer values. """ res = {} if counters and config.get_config()["counters_backend"] == "swh-counters": res = counters.get_counts( ["origin", "revision", "content", "directory", "release", "person"] ) else: res = storage.stat_counters() return res def _lookup_origin_visits( origin_url: str, last_visit: Optional[int] = None, limit: int = 10 ) -> Iterator[OriginVisit]: """Yields the origin origins' visits. Args: origin_url (str): origin to list visits for last_visit (int): last visit to lookup from limit (int): Number of elements max to display Yields: OriginVisit for that origin """ limit = min(limit, MAX_LIMIT) page_token: Optional[str] if last_visit is not None: page_token = str(last_visit) else: page_token = None visit_page = storage.origin_visit_get( origin_url, page_token=page_token, limit=limit ) yield from visit_page.results def lookup_origin_visits( origin: str, last_visit: Optional[int] = None, per_page: int = 10 ) -> Iterator[OriginVisitInfo]: """Yields the origin origins' visits. Args: origin: origin to list visits for Yields: Dictionaries of origin_visit for that origin """ for visit in _lookup_origin_visits(origin, last_visit=last_visit, limit=per_page): visit_status = storage.origin_visit_status_get_latest(origin, visit.visit) yield converters.from_origin_visit( {**visit_status.to_dict(), "type": visit.type} ) def lookup_origin_visit_latest( origin_url: str, require_snapshot: bool = False, type: Optional[str] = None, allowed_statuses: Optional[List[str]] = None, ) -> Optional[OriginVisitInfo]: """Return the origin's latest visit Args: origin_url: origin to list visits for type: Optional visit type to filter on (e.g git, tar, dsc, svn, hg, npm, pypi, ...) allowed_statuses: list of visit statuses considered to find the latest visit. For instance, ``allowed_statuses=['full']`` will only consider visits that have successfully run to completion. require_snapshot: filter out origins without a snapshot Returns: The origin visit info as dict if found """ visit_status = origin_get_latest_visit_status( storage, origin_url, type=type, allowed_statuses=allowed_statuses, require_snapshot=require_snapshot, ) return ( converters.from_origin_visit(visit_status.to_dict()) if visit_status else None ) def lookup_origin_visit(origin_url: str, visit_id: int) -> OriginVisitInfo: """Return information about visit visit_id with origin origin. Args: origin: origin concerned by the visit visit_id: the visit identifier to lookup Yields: The dict origin_visit concerned """ visit = storage.origin_visit_get_by(origin_url, visit_id) visit_status = storage.origin_visit_status_get_latest(origin_url, visit_id) if not visit: raise NotFoundExc( f"Origin {origin_url} or its visit with id {visit_id} not found!" ) return converters.from_origin_visit({**visit_status.to_dict(), "type": visit.type}) def lookup_snapshot_sizes( snapshot_id: str, branch_name_exclude_prefix: Optional[str] = "refs/pull/" ) -> Dict[str, int]: """Count the number of branches in the snapshot with the given id Args: snapshot_id (str): sha1 identifier of the snapshot Returns: dict: A dict whose keys are the target types of branches and values their corresponding amount """ snapshot_id_bin = _to_sha1_bin(snapshot_id) snapshot_sizes = dict.fromkeys(("alias", "release", "revision"), 0) branch_counts = storage.snapshot_count_branches( snapshot_id_bin, branch_name_exclude_prefix.encode() if branch_name_exclude_prefix else None, ) # remove possible None key returned by snapshot_count_branches # when null branches are present in the snapshot branch_counts.pop(None, None) snapshot_sizes.update(branch_counts) return snapshot_sizes def lookup_snapshot( snapshot_id: str, branches_from: str = "", branches_count: int = 1000, target_types: Optional[List[str]] = None, branch_name_include_substring: Optional[str] = None, branch_name_exclude_prefix: Optional[str] = "refs/pull/", ) -> Dict[str, Any]: """Return information about a snapshot, aka the list of named branches found during a specific visit of an origin. Args: snapshot_id: sha1 identifier of the snapshot branches_from: optional parameter used to skip branches whose name is lesser than it before returning them branches_count: optional parameter used to restrain the amount of returned branches target_types: optional parameter used to filter the target types of branch to return (possible values that can be contained in that list are `'content', 'directory', 'revision', 'release', 'snapshot', 'alias'`) branch_name_include_substring: if provided, only return branches whose name contains given substring branch_name_exclude_prefix: if provided, do not return branches whose name starts with given pattern Raises: NotFoundExc if the given snapshot_id is missing Returns: A dict filled with the snapshot content. """ snapshot_id_bin = _to_sha1_bin(snapshot_id) if storage.snapshot_missing([snapshot_id_bin]): raise NotFoundExc(f"Snapshot with id {snapshot_id} not found!") partial_branches = storage.snapshot_get_branches( snapshot_id_bin, branches_from.encode(), branches_count, target_types, branch_name_include_substring.encode() if branch_name_include_substring else None, branch_name_exclude_prefix.encode() if branch_name_exclude_prefix else None, ) return ( converters.from_partial_branches(partial_branches) if partial_branches else None ) def lookup_latest_origin_snapshot( origin: str, allowed_statuses: List[str] = None ) -> Optional[Dict[str, Any]]: """Return information about the latest snapshot of an origin. .. warning:: At most 1000 branches contained in the snapshot will be returned for performance reasons. Args: origin: URL or integer identifier of the origin allowed_statuses: list of visit statuses considered to find the latest snapshot for the visit. For instance, ``allowed_statuses=['full']`` will only consider visits that have successfully run to completion. Returns: A dict filled with the snapshot content. """ snp = snapshot_get_latest( storage, origin, allowed_statuses=allowed_statuses, branches_count=1000 ) return converters.from_snapshot(snp.to_dict()) if snp is not None else None def lookup_snapshot_alias( snapshot_id: str, alias_name: str ) -> Optional[Dict[str, Any]]: """Try to resolve a branch alias in a snapshot. Args: snapshot_id: hexadecimal representation of a snapshot id alias_name: name of the branch alias to resolve Returns: Target branch information or None if the alias does not exist or target a dangling branch. """ resolved_alias = snapshot_resolve_alias( storage, _to_sha1_bin(snapshot_id), alias_name.encode() ) return ( converters.from_swh(resolved_alias.to_dict(), hashess={"target"}) if resolved_alias is not None else None ) def lookup_revision_through(revision, limit=100): """Retrieve a revision from the criterion stored in revision dictionary. Args: revision: Dictionary of criterion to lookup the revision with. Here are the supported combination of possible values: - origin_url, branch_name, ts, sha1_git - origin_url, branch_name, ts - sha1_git_root, sha1_git - sha1_git Returns: None if the revision is not found or the actual revision. """ if ( "origin_url" in revision and "branch_name" in revision and "ts" in revision and "sha1_git" in revision ): return lookup_revision_with_context_by( revision["origin_url"], revision["branch_name"], revision["ts"], revision["sha1_git"], limit, ) if "origin_url" in revision and "branch_name" in revision and "ts" in revision: return lookup_revision_by( revision["origin_url"], revision["branch_name"], revision["ts"] ) if "sha1_git_root" in revision and "sha1_git" in revision: return lookup_revision_with_context( revision["sha1_git_root"], revision["sha1_git"], limit ) if "sha1_git" in revision: return lookup_revision(revision["sha1_git"]) # this should not happen raise NotImplementedError("Should not happen!") def lookup_directory_through_revision(revision, path=None, limit=100, with_data=False): """Retrieve the directory information from the revision. Args: revision: dictionary of criterion representing a revision to lookup path: directory's path to lookup. limit: optional query parameter to limit the revisions log (default to 100). For now, note that this limit could impede the transitivity conclusion about sha1_git not being an ancestor of. with_data: indicate to retrieve the content's raw data if path resolves to a content. Returns: The directory pointing to by the revision criterions at path. """ rev = lookup_revision_through(revision, limit) if not rev: raise NotFoundExc("Revision with criterion %s not found!" % revision) return (rev["id"], lookup_directory_with_revision(rev["id"], path, with_data)) def _vault_request(vault_fn, bundle_type: str, swhid: CoreSWHID, **kwargs): try: return vault_fn(bundle_type, swhid, **kwargs) except VaultNotFoundExc: return None def vault_cook(bundle_type: str, swhid: CoreSWHID, email=None): """Cook a vault bundle. """ return _vault_request(vault.cook, bundle_type, swhid, email=email) def vault_fetch(bundle_type: str, swhid: CoreSWHID): """Fetch a vault bundle. """ return _vault_request(vault.fetch, bundle_type, swhid) def vault_progress(bundle_type: str, swhid: CoreSWHID): """Get the current progress of a vault bundle. """ return _vault_request(vault.progress, bundle_type, swhid) def diff_revision(rev_id): """Get the list of file changes (insertion / deletion / modification / renaming) for a particular revision. """ rev_sha1_git_bin = _to_sha1_bin(rev_id) changes = diff.diff_revision(storage, rev_sha1_git_bin, track_renaming=True) for change in changes: change["from"] = converters.from_directory_entry(change["from"]) change["to"] = converters.from_directory_entry(change["to"]) if change["from_path"]: change["from_path"] = change["from_path"].decode("utf-8") if change["to_path"]: change["to_path"] = change["to_path"].decode("utf-8") return changes class _RevisionsWalkerProxy(object): """ Proxy class wrapping a revisions walker iterator from swh-storage and performing needed conversions. """ def __init__(self, rev_walker_type, rev_start, *args, **kwargs): rev_start_bin = hashutil.hash_to_bytes(rev_start) self.revisions_walker = revisions_walker.get_revisions_walker( rev_walker_type, storage, rev_start_bin, *args, **kwargs ) def export_state(self): return self.revisions_walker.export_state() def __next__(self): return converters.from_revision(next(self.revisions_walker)) def __iter__(self): return self def get_revisions_walker(rev_walker_type, rev_start, *args, **kwargs): """ Utility function to instantiate a revisions walker of a given type, see :mod:`swh.storage.algos.revisions_walker`. Args: rev_walker_type (str): the type of revisions walker to return, possible values are: ``committer_date``, ``dfs``, ``dfs_post``, ``bfs`` and ``path`` rev_start (str): hexadecimal representation of a revision identifier args (list): position arguments to pass to the revisions walker constructor kwargs (dict): keyword arguments to pass to the revisions walker constructor """ # first check if the provided revision is valid lookup_revision(rev_start) return _RevisionsWalkerProxy(rev_walker_type, rev_start, *args, **kwargs) def lookup_object(object_type: ObjectType, object_id: str) -> Dict[str, Any]: """ Utility function for looking up an object in the archive by its type and id. Args: object_type (str): the type of object to lookup, either *content*, *directory*, *release*, *revision* or *snapshot* object_id (str): the *sha1_git* checksum identifier in hexadecimal form of the object to lookup Returns: Dict[str, Any]: A dictionary describing the object or a list of dictionary for the directory object type. Raises: swh.web.common.exc.NotFoundExc: if the object could not be found in the archive BadInputExc: if the object identifier is invalid """ if object_type == ObjectType.CONTENT: return lookup_content(f"sha1_git:{object_id}") elif object_type == ObjectType.DIRECTORY: return {"id": object_id, "content": list(lookup_directory(object_id))} elif object_type == ObjectType.RELEASE: return lookup_release(object_id) elif object_type == ObjectType.REVISION: return lookup_revision(object_id) elif object_type == ObjectType.SNAPSHOT: return lookup_snapshot(object_id) else: raise ValueError(f"Unexpected object type variant: {object_type}") def lookup_missing_hashes(grouped_swhids: Dict[str, List[bytes]]) -> Set[str]: """Lookup missing Software Heritage persistent identifier hash, using batch processing. Args: A dictionary with: keys: object types values: object hashes Returns: A set(hexadecimal) of the hashes not found in the storage """ missing_hashes = [] for obj_type, obj_ids in grouped_swhids.items(): if obj_type == ObjectType.CONTENT: missing_hashes.append(storage.content_missing_per_sha1_git(obj_ids)) elif obj_type == ObjectType.DIRECTORY: missing_hashes.append(storage.directory_missing(obj_ids)) elif obj_type == ObjectType.REVISION: missing_hashes.append(storage.revision_missing(obj_ids)) elif obj_type == ObjectType.RELEASE: missing_hashes.append(storage.release_missing(obj_ids)) elif obj_type == ObjectType.SNAPSHOT: missing_hashes.append(storage.snapshot_missing(obj_ids)) missing = set( map(lambda x: hashutil.hash_to_hex(x), itertools.chain(*missing_hashes)) ) return missing def lookup_origins_by_sha1s(sha1s: List[str]) -> Iterator[Optional[OriginInfo]]: """Lookup origins from the sha1 hash values of their URLs. Args: sha1s: list of sha1s hexadecimal representation Yields: origin information as dict """ sha1s_bytes = [hashutil.hash_to_bytes(sha1) for sha1 in sha1s] origins = storage.origin_get_by_sha1(sha1s_bytes) for origin in origins: yield converters.from_origin(origin) diff --git a/swh/web/common/converters.py b/swh/web/common/converters.py index cdf1a6a3..2deecfcf 100644 --- a/swh/web/common/converters.py +++ b/swh/web/common/converters.py @@ -1,409 +1,409 @@ # Copyright (C) 2015-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import datetime import json from typing import Any, Dict, Union from django.core.serializers.json import DjangoJSONEncoder from swh.core.utils import decode_with_escape from swh.model import hashutil -from swh.model.identifiers import ObjectType from swh.model.model import RawExtrinsicMetadata, Release, Revision +from swh.model.swhids import ObjectType from swh.storage.interface import PartialBranches from swh.web.common.typing import OriginInfo, OriginVisitInfo def _group_checksums(data): """Groups checksums values computed from hash functions used in swh and stored in data dict under a single entry 'checksums' """ if data: checksums = {} for hash in hashutil.ALGORITHMS: if hash in data and data[hash]: checksums[hash] = data[hash] del data[hash] if len(checksums) > 0: data["checksums"] = checksums def fmap(f, data): """Map f to data at each level. This must keep the origin data structure type: - map -> map - dict -> dict - list -> list - None -> None Args: f: function that expects one argument. data: data to traverse to apply the f function. list, map, dict or bare value. Returns: The same data-structure with modified values by the f function. """ if data is None: return data if isinstance(data, map): return map(lambda y: fmap(f, y), (x for x in data)) if isinstance(data, list): return [fmap(f, x) for x in data] if isinstance(data, tuple): return tuple(fmap(f, x) for x in data) if isinstance(data, dict): return {k: fmap(f, v) for (k, v) in data.items()} return f(data) def from_swh( dict_swh, hashess={}, bytess={}, dates={}, blacklist={}, removables_if_empty={}, empty_dict={}, empty_list={}, convert={}, convert_fn=lambda x: x, ): """Convert from a swh dictionary to something reasonably json serializable. Args: dict_swh: the origin dictionary needed to be transformed hashess: list/set of keys representing hashes values (sha1, sha256, sha1_git, etc...) as bytes. Those need to be transformed in hexadecimal string bytess: list/set of keys representing bytes values which needs to be decoded blacklist: set of keys to filter out from the conversion convert: set of keys whose associated values need to be converted using convert_fn convert_fn: the conversion function to apply on the value of key in 'convert' The remaining keys are copied as is in the output. Returns: dictionary equivalent as dict_swh only with its keys converted. """ def convert_hashes_bytes(v): """v is supposedly a hash as bytes, returns it converted in hex. """ if isinstance(v, bytes): return hashutil.hash_to_hex(v) return v def convert_bytes(v): """v is supposedly a bytes string, decode as utf-8. FIXME: Improve decoding policy. If not utf-8, break! """ if isinstance(v, bytes): return v.decode("utf-8") return v def convert_date(v): """ Args: v (dict or datatime): either: - a dict with three keys: - timestamp (dict or integer timestamp) - offset - negative_utc - or, a datetime We convert it to a human-readable string """ if not v: return v if isinstance(v, datetime.datetime): return v.isoformat() tz = datetime.timezone(datetime.timedelta(minutes=v["offset"])) swh_timestamp = v["timestamp"] if isinstance(swh_timestamp, dict): date = datetime.datetime.fromtimestamp(swh_timestamp["seconds"], tz=tz) else: date = datetime.datetime.fromtimestamp(swh_timestamp, tz=tz) datestr = date.isoformat() if v["offset"] == 0 and v["negative_utc"]: # remove the rightmost + and replace it with a - return "-".join(datestr.rsplit("+", 1)) return datestr if not dict_swh: return dict_swh new_dict = {} for key, value in dict_swh.items(): if key in blacklist or (key in removables_if_empty and not value): continue if key in dates: new_dict[key] = convert_date(value) elif key in convert: new_dict[key] = convert_fn(value) elif isinstance(value, dict): new_dict[key] = from_swh( value, hashess=hashess, bytess=bytess, dates=dates, blacklist=blacklist, removables_if_empty=removables_if_empty, empty_dict=empty_dict, empty_list=empty_list, convert=convert, convert_fn=convert_fn, ) elif key in hashess: new_dict[key] = fmap(convert_hashes_bytes, value) elif key in bytess: try: new_dict[key] = fmap(convert_bytes, value) except UnicodeDecodeError: if "decoding_failures" not in new_dict: new_dict["decoding_failures"] = [key] else: new_dict["decoding_failures"].append(key) new_dict[key] = fmap(decode_with_escape, value) elif key in empty_dict and not value: new_dict[key] = {} elif key in empty_list and not value: new_dict[key] = [] else: new_dict[key] = value _group_checksums(new_dict) return new_dict def from_origin(origin: Dict[str, Any]) -> OriginInfo: """Convert from a swh origin to an origin dictionary. """ return from_swh(origin, blacklist={"id"}) def from_release(release: Release) -> Dict[str, Any]: """Convert from a swh release to a json serializable release dictionary. Args: release: A release model object Returns: release dictionary with the following keys - id: hexadecimal sha1 (string) - revision: hexadecimal sha1 (string) - comment: release's comment message (string) - name: release's name (string) - author: release's author identifier (swh's id) - synthetic: the synthetic property (boolean) """ return from_swh( release.to_dict(), hashess={"id", "target"}, bytess={"message", "name", "fullname", "email"}, dates={"date"}, ) class SWHDjangoJSONEncoder(DjangoJSONEncoder): """Wrapper around DjangoJSONEncoder to serialize SWH-specific types found in :class:`swh.web.common.typing.SWHObjectInfo`.""" def default(self, o): if isinstance(o, ObjectType): return o.name.lower() else: super().default(o) class SWHMetadataEncoder(json.JSONEncoder): """Special json encoder for metadata field which can contain bytes encoded value. """ def default(self, obj): if isinstance(obj, bytes): try: return obj.decode("utf-8") except UnicodeDecodeError: # fallback to binary representation to avoid display errors return repr(obj) # Let the base class default method raise the TypeError return json.JSONEncoder.default(self, obj) def convert_metadata(metadata): """Convert json specific dict to a json serializable one. """ if metadata is None: return {} return json.loads(json.dumps(metadata, cls=SWHMetadataEncoder)) def from_revision(revision: Union[Dict[str, Any], Revision]) -> Dict[str, Any]: """Convert swh revision model object to a json serializable revision dictionary. Args: revision: revision model object Returns: dict: Revision dictionary with the same keys as inputs, except: - sha1s are in hexadecimal strings (id, directory) - bytes are decoded in string (author_name, committer_name, author_email, committer_email) Remaining keys are left as is """ if isinstance(revision, Revision): revision_d = revision.to_dict() else: revision_d = revision revision_d = from_swh( revision_d, hashess={"id", "directory", "parents", "children"}, bytess={"name", "fullname", "email", "extra_headers", "message"}, convert={"metadata"}, convert_fn=convert_metadata, dates={"date", "committer_date"}, ) if revision_d: if "parents" in revision_d: revision_d["merge"] = len(revision_d["parents"]) > 1 return revision_d def from_raw_extrinsic_metadata( metadata: Union[Dict[str, Any], RawExtrinsicMetadata] ) -> Dict[str, Any]: """Convert RawExtrinsicMetadata model object to a json serializable dictionary. """ return from_swh( metadata.to_dict() if isinstance(metadata, RawExtrinsicMetadata) else metadata, blacklist={"id", "metadata"}, dates={"discovery_date"}, ) def from_content(content): """Convert swh content to serializable content dictionary. """ return from_swh( content, hashess={"sha1", "sha1_git", "sha256", "blake2s256"}, blacklist={"ctime"}, convert={"status"}, convert_fn=lambda v: "absent" if v == "hidden" else v, ) def from_person(person): """Convert swh person to serializable person dictionary. """ return from_swh(person, bytess={"name", "fullname", "email"}) def from_origin_visit(visit: Dict[str, Any]) -> OriginVisitInfo: """Convert swh origin_visit to serializable origin_visit dictionary. """ ov = from_swh( visit, hashess={"target", "snapshot"}, bytess={"branch"}, dates={"date"}, empty_dict={"metadata"}, ) return ov def from_snapshot(snapshot): """Convert swh snapshot to serializable (partial) snapshot dictionary. """ sv = from_swh(snapshot, hashess={"id", "target"}, bytess={"next_branch"}) if sv and "branches" in sv: sv["branches"] = {decode_with_escape(k): v for k, v in sv["branches"].items()} for k, v in snapshot["branches"].items(): # alias target existing branch names, not a sha1 if v and v["target_type"] == "alias": branch = decode_with_escape(k) target = decode_with_escape(v["target"]) sv["branches"][branch]["target"] = target return sv def from_partial_branches(branches: PartialBranches): """Convert PartialBranches to serializable partial snapshot dictionary """ return from_snapshot( { "id": branches["id"], "branches": { branch_name: branch.to_dict() if branch else None for (branch_name, branch) in branches["branches"].items() }, "next_branch": branches["next_branch"], } ) def from_directory_entry(dir_entry): """Convert swh directory to serializable directory dictionary. """ return from_swh( dir_entry, hashess={"dir_id", "sha1_git", "sha1", "sha256", "blake2s256", "target"}, bytess={"name"}, removables_if_empty={"sha1", "sha1_git", "sha256", "blake2s256", "status"}, convert={"status"}, convert_fn=lambda v: "absent" if v == "hidden" else v, ) def from_filetype(content_entry): """Convert swh content to serializable dictionary containing keys 'id', 'encoding', and 'mimetype'. """ return from_swh(content_entry, hashess={"id"}) diff --git a/swh/web/common/identifiers.py b/swh/web/common/identifiers.py index 43866f76..d3492909 100644 --- a/swh/web/common/identifiers.py +++ b/swh/web/common/identifiers.py @@ -1,388 +1,388 @@ # Copyright (C) 2020-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from typing import Any, Dict, Iterable, List, Optional from urllib.parse import quote, unquote from typing_extensions import TypedDict from django.http import QueryDict from swh.model.exceptions import ValidationError from swh.model.hashutil import hash_to_bytes, hash_to_hex -from swh.model.identifiers import ObjectType, QualifiedSWHID +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] = 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"] = query_dict["path"][1:] elif object_type == ObjectType.DIRECTORY: object_id = directory # remove leading and trailing slashes from SWHID directory path if query_dict["path"].endswith("/"): query_dict["path"] = query_dict["path"][1:-1] else: query_dict["path"] = query_dict["path"][1:] # snapshot context if swhid_parsed.visit: if swhid_parsed.visit.object_type != ObjectType.SNAPSHOT: raise BadInputExc("Visit must be a snapshot SWHID.") query_dict["snapshot"] = hash_to_hex(swhid_parsed.visit.object_id) if swhid_parsed.anchor: if ( swhid_parsed.anchor.object_type == ObjectType.REVISION and object_type != ObjectType.REVISION ): query_dict["revision"] = hash_to_hex(swhid_parsed.anchor.object_id) elif swhid_parsed.anchor.object_type == ObjectType.RELEASE: release = archive.lookup_release( hash_to_hex(swhid_parsed.anchor.object_id) ) if release: query_dict["release"] = release["name"] # browsing content or directory without snapshot context elif ( object_type in (ObjectType.CONTENT, ObjectType.DIRECTORY) and swhid_parsed.anchor ): if swhid_parsed.anchor.object_type == ObjectType.REVISION: # anchor revision, objects are browsed from its view object_type = ObjectType.REVISION object_id = swhid_parsed.anchor.object_id elif ( object_type == ObjectType.DIRECTORY and swhid_parsed.anchor.object_type == ObjectType.DIRECTORY ): # a directory is browsed from its root object_id = swhid_parsed.anchor.object_id if object_type == ObjectType.CONTENT: url_args["query_string"] = f"sha1_git:{hash_to_hex(object_id)}" elif object_type in (ObjectType.DIRECTORY, ObjectType.RELEASE, ObjectType.REVISION): url_args["sha1_git"] = hash_to_hex(object_id) elif object_type == ObjectType.SNAPSHOT: url_args["snapshot_id"] = hash_to_hex(object_id) if swhid_parsed.lines and process_lines: lines = swhid_parsed.lines fragment += "#L" + str(lines[0]) if lines[1]: fragment += "-L" + str(lines[1]) if url_args: browse_url = ( reverse( f"browse-{object_type.name.lower()}", url_args=url_args, query_params=query_dict, ) + fragment ) return ResolvedSWHID(swhid_parsed=swhid_parsed, browse_url=browse_url) def get_swhid(swhid: str) -> QualifiedSWHID: """Check if a SWHID is valid and return it parsed. Args: swhid: a SoftWare Heritage persistent IDentifier. Raises: BadInputExc: if the provided SWHID can not be parsed. Return: A parsed SWHID. """ try: # ensure core part of SWHID is in lower case to avoid parsing error (core, sep, qualifiers) = swhid.partition(";") core = core.lower() return QualifiedSWHID.from_string(core + sep + qualifiers) except ValidationError as ve: raise BadInputExc("Error when parsing identifier: %s" % " ".join(ve.messages)) def group_swhids(swhids: Iterable[QualifiedSWHID],) -> Dict[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, ) -> 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 e9cad745..04e1616a 100644 --- a/swh/web/common/typing.py +++ b/swh/web/common/typing.py @@ -1,261 +1,261 @@ # Copyright (C) 2020-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from typing import Any, Dict, List, Optional, TypeVar, Union from typing_extensions import TypedDict from django.http import QueryDict from swh.core.api.classes import PagedResult as CorePagedResult -from swh.model.identifiers import ObjectType +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 associatd to the release""" id: str """release identifier""" message: str """release message""" name: str """release name""" alias: bool """define if the branch is an alias""" target: str """release target""" target_type: str """release target_type""" url: Optional[str] """optional browse URL (content, directory, ...) scoped to release""" class SnapshotContext(TypedDict): branch: Optional[str] """optional branch name set when browsing snapshot in that scope""" branch_alias: bool """indicates if the focused branch is an alias""" branches: List[SnapshotBranchInfo] """list of snapshot branches (possibly truncated)""" branches_url: str """snapshot branches list browse URL""" is_empty: bool """indicates if the snapshot is empty""" origin_info: Optional[OriginInfo] """optional origin info associated to the snapshot""" origin_visits_url: Optional[str] """optional origin visits URL""" query_params: QueryParameters """common query parameters when browsing snapshot content""" release: Optional[str] """optional release name set when browsing snapshot in that scope""" release_alias: bool """indicates if the focused release is an alias""" release_id: Optional[str] """optional release identifier set when browsing snapshot in that scope""" releases: List[SnapshotReleaseInfo] """list of snapshot releases (possibly truncated)""" releases_url: str """snapshot releases list browse URL""" revision_id: Optional[str] """optional revision identifier set when browsing snapshot in that scope""" revision_info: Optional[Dict[str, Any]] """optional revision info set when browsing snapshot in that scope""" root_directory: Optional[str] """optional root directory identifier set when browsing snapshot content""" snapshot_id: str """snapshot identifier""" snapshot_sizes: Dict[str, int] """snapshot sizes grouped by branch target type""" snapshot_swhid: str """snapshot SWHID""" url_args: Dict[str, Any] """common URL arguments when browsing snapshot content""" visit_info: Optional[OriginVisitInfo] """optional origin visit info associated to the snapshot""" class SWHObjectInfo(TypedDict): object_type: ObjectType object_id: str class SWHIDContext(TypedDict, total=False): origin: str anchor: str visit: str path: str lines: str class SWHIDInfo(SWHObjectInfo): swhid: str swhid_url: str context: SWHIDContext swhid_with_context: Optional[str] swhid_with_context_url: Optional[str] class SWHObjectInfoMetadata(TypedDict, total=False): origin_url: Optional[str] visit_date: Optional[str] visit_type: Optional[str] class ContentMetadata(SWHObjectInfo, SWHObjectInfoMetadata): sha1: str sha1_git: str sha256: str blake2s256: str content_url: str mimetype: str encoding: str size: int language: str path: Optional[str] filename: Optional[str] directory: Optional[str] root_directory: Optional[str] revision: Optional[str] release: Optional[str] snapshot: Optional[str] class DirectoryMetadata(SWHObjectInfo, SWHObjectInfoMetadata): directory: str nb_files: int nb_dirs: int sum_file_sizes: int root_directory: Optional[str] path: str revision: Optional[str] revision_found: Optional[bool] release: Optional[str] snapshot: Optional[str] class ReleaseMetadata(SWHObjectInfo, SWHObjectInfoMetadata): release: str author: str author_url: str date: str name: str synthetic: bool target: str target_type: str snapshot: Optional[str] class RevisionMetadata(SWHObjectInfo, SWHObjectInfoMetadata): revision: str author: str author_url: str committer: str committer_url: str date: str committer_date: str directory: str merge: bool metadata: str parents: List[str] synthetic: bool type: str snapshot: Optional[str] TResult = TypeVar("TResult") PagedResult = CorePagedResult[TResult, str] class SaveOriginRequestInfo(TypedDict): id: int """Unique key""" save_request_date: str """Date of the creation request""" visit_type: str """Type of the visit""" visit_status: Optional[str] """Status of the visit""" origin_url: str """Origin to ingest""" save_request_status: str """Status of the request""" loading_task_id: Optional[int] """Identifier of the loading task in the scheduler if scheduled""" visit_date: Optional[str] """End of the visit if terminated""" save_task_status: str """Status of the scheduled task""" 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)""" diff --git a/swh/web/misc/badges.py b/swh/web/misc/badges.py index a6e30d3e..c5548f32 100644 --- a/swh/web/misc/badges.py +++ b/swh/web/misc/badges.py @@ -1,168 +1,168 @@ # Copyright (C) 2019-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from base64 import b64encode from typing import Optional, cast from pybadges import badge from django.conf.urls import url from django.contrib.staticfiles import finders from django.http import HttpRequest, HttpResponse from swh.model.exceptions import ValidationError from swh.model.hashutil import hash_to_bytes, hash_to_hex -from swh.model.identifiers import CoreSWHID, ObjectType, QualifiedSWHID +from swh.model.swhids import CoreSWHID, ObjectType, QualifiedSWHID from swh.web.common import archive from swh.web.common.exc import BadInputExc, NotFoundExc from swh.web.common.identifiers import parse_object_type, resolve_swhid from swh.web.common.utils import reverse _orange = "#f36a24" _blue = "#0172b2" _red = "#cd5741" _swh_logo_data = None _badge_config = { "content": {"color": _blue, "title": "Archived source file",}, "directory": {"color": _blue, "title": "Archived source tree",}, "origin": {"color": _orange, "title": "Archived software repository",}, "release": {"color": _blue, "title": "Archived software release",}, "revision": {"color": _blue, "title": "Archived commit",}, "snapshot": {"color": _blue, "title": "Archived software repository snapshot",}, "error": {"color": _red, "title": "An error occurred when generating the badge"}, } def _get_logo_data() -> str: """ Get data-URI for Software Heritage SVG logo to embed it in the generated badges. """ global _swh_logo_data if _swh_logo_data is None: swh_logo_path = cast(str, finders.find("img/swh-logo-white.svg")) with open(swh_logo_path, "rb") as swh_logo_file: _swh_logo_data = "data:image/svg+xml;base64,%s" % b64encode( swh_logo_file.read() ).decode("ascii") return _swh_logo_data def _swh_badge( request: HttpRequest, object_type: str, object_id: str, object_swhid: Optional[str] = "", ) -> HttpResponse: """ Generate a Software Heritage badge for a given object type and id. Args: request: input http request object_type: The type of swh object to generate a badge for, either *content*, *directory*, *revision*, *release*, *origin* or *snapshot* object_id: The id of the swh object, either an url for origin type or a *sha1* for other object types object_swhid: If provided, the object SWHID will not be recomputed Returns: HTTP response with content type *image/svg+xml* containing the SVG badge data. If the provided parameters are invalid, HTTP 400 status code will be returned. If the object can not be found in the archive, HTTP 404 status code will be returned. """ left_text = "error" whole_link = None try: if object_type == "origin": archive.lookup_origin({"url": object_id}) right_text = "repository" whole_link = reverse( "browse-origin", query_params={"origin_url": object_id} ) else: # when SWHID is provided, object type and id will be parsed # from it if object_swhid: parsed_swhid = QualifiedSWHID.from_string(object_swhid) parsed_object_type = parsed_swhid.object_type object_id = hash_to_hex(parsed_swhid.object_id) swh_object = archive.lookup_object(parsed_swhid.object_type, object_id) # remove SWHID qualified if any for badge text right_text = str( CoreSWHID( object_type=parsed_swhid.object_type, object_id=parsed_swhid.object_id, ) ) object_type = parsed_swhid.object_type.name.lower() else: parsed_object_type = parse_object_type(object_type) right_text = str( CoreSWHID( object_type=parsed_object_type, object_id=hash_to_bytes(object_id), ) ) swh_object = archive.lookup_object(parsed_object_type, object_id) whole_link = resolve_swhid(str(right_text))["browse_url"] # use release name for badge text if parsed_object_type == ObjectType.RELEASE: right_text = "release %s" % swh_object["name"] left_text = "archived" except (BadInputExc, ValidationError): right_text = f'invalid {object_type if object_type else "object"} id' object_type = "error" except NotFoundExc: right_text = f'{object_type if object_type else "object"} not found' object_type = "error" badge_data = badge( left_text=left_text, right_text=right_text, right_color=_badge_config[object_type]["color"], whole_link=request.build_absolute_uri(whole_link), whole_title=_badge_config[object_type]["title"], logo=_get_logo_data(), embed_logo=True, ) return HttpResponse(badge_data, content_type="image/svg+xml") def _swh_badge_swhid(request: HttpRequest, object_swhid: str) -> HttpResponse: """ Generate a Software Heritage badge for a given object SWHID. Args: request (django.http.HttpRequest): input http request object_swhid (str): a SWHID of an archived object Returns: django.http.HttpResponse: An http response with content type *image/svg+xml* containing the SVG badge data. If any error occurs, a status code of 400 will be returned. """ return _swh_badge(request, "", "", object_swhid) urlpatterns = [ url( r"^badge/(?P[a-z]+)/(?P.+)/$", _swh_badge, name="swh-badge", ), url( r"^badge/(?Pswh:[0-9]+:[a-z]+:[0-9a-f]+.*)/$", _swh_badge_swhid, name="swh-badge-swhid", ), ] diff --git a/swh/web/misc/iframe.py b/swh/web/misc/iframe.py index 5c19f98c..83f2f697 100644 --- a/swh/web/misc/iframe.py +++ b/swh/web/misc/iframe.py @@ -1,306 +1,306 @@ # Copyright (C) 2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from typing import Any, Dict, List, Optional from django.conf.urls import url from django.shortcuts import render from django.views.decorators.clickjacking import xframe_options_exempt from swh.model.hashutil import hash_to_bytes -from swh.model.identifiers import ObjectType, QualifiedSWHID +from swh.model.swhids import ObjectType, QualifiedSWHID from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.browse.utils import ( content_display_max_size, get_directory_entries, prepare_content_for_display, request_content, ) from swh.web.common import archive from swh.web.common.exc import BadInputExc, NotFoundExc, http_status_code_message from swh.web.common.identifiers import get_swhid, get_swhids_info from swh.web.common.typing import SnapshotContext, SWHObjectInfo from swh.web.common.utils import gen_path_info, reverse def _get_content_rendering_data(cnt_swhid: QualifiedSWHID, path: str) -> Dict[str, Any]: content_data = request_content(f"sha1_git:{cnt_swhid.object_id.hex()}") content = None language = None mimetype = None if content_data.get("raw_data") is not None: content_display_data = prepare_content_for_display( content_data["raw_data"], content_data["mimetype"], path ) content = content_display_data["content_data"] language = content_display_data["language"] mimetype = content_display_data["mimetype"] return { "content": content, "content_size": content_data.get("length"), "max_content_size": content_display_max_size, "filename": path.split("/")[-1], "encoding": content_data.get("encoding"), "mimetype": mimetype, "language": language, } def _get_directory_rendering_data( dir_swhid: QualifiedSWHID, focus_swhid: QualifiedSWHID, path: str, ) -> Dict[str, Any]: dirs, files = get_directory_entries(dir_swhid.object_id.hex()) for d in dirs: if d["type"] == "rev": d["url"] = None else: dir_swhid = QualifiedSWHID( object_type=ObjectType.DIRECTORY, object_id=hash_to_bytes(d["target"]), origin=dir_swhid.origin, visit=dir_swhid.visit, anchor=dir_swhid.anchor, path=(path or "/") + d["name"] + "/", ) d["url"] = reverse( "swhid-iframe", url_args={"swhid": str(dir_swhid)}, query_params={"focus_swhid": str(focus_swhid)}, ) for f in files: object_id = hash_to_bytes(f["target"]) cnt_swhid = QualifiedSWHID( object_type=ObjectType.CONTENT, object_id=object_id, origin=dir_swhid.origin, visit=dir_swhid.visit, anchor=dir_swhid.anchor, path=(path or "/") + f["name"], lines=(focus_swhid.lines if object_id == focus_swhid.object_id else None), ) f["url"] = reverse( "swhid-iframe", url_args={"swhid": str(cnt_swhid)}, query_params={"focus_swhid": str(focus_swhid)}, ) return {"dirs": dirs, "files": files} def _get_breacrumbs_data( swhid: QualifiedSWHID, focus_swhid: QualifiedSWHID, path: str, snapshot_context: Optional[SnapshotContext] = None, ) -> List[Dict[str, Any]]: breadcrumbs = [] filename = None # strip any leading or trailing slash from path qualifier of SWHID if path and path[0] == "/": path = path[1:] if path and path[-1] == "/": path = path[:-1] if swhid.object_type == ObjectType.CONTENT: split_path = path.split("/") filename = split_path[-1] path = path[: -len(filename)] path_info = gen_path_info(path) if path != "/" else [] root_dir = None if snapshot_context and snapshot_context["root_directory"]: root_dir = snapshot_context["root_directory"] elif focus_swhid.object_type == ObjectType.DIRECTORY: root_dir = focus_swhid.object_id.hex() if root_dir: root_dir_swhid = QualifiedSWHID( object_type=ObjectType.DIRECTORY, object_id=hash_to_bytes(root_dir), origin=swhid.origin, visit=swhid.visit, anchor=swhid.anchor, ) breadcrumbs.append( { "name": root_dir[:7], "object_id": root_dir_swhid.object_id.hex(), "path": "/", "url": reverse( "swhid-iframe", url_args={"swhid": str(root_dir_swhid)}, query_params={"focus_swhid": focus_swhid}, ), } ) for pi in path_info: dir_info = archive.lookup_directory_with_path(root_dir, pi["path"]) dir_swhid = QualifiedSWHID( object_type=ObjectType.DIRECTORY, object_id=hash_to_bytes(dir_info["target"]), origin=swhid.origin, visit=swhid.visit, anchor=swhid.anchor, path="/" + pi["path"] + "/", ) breadcrumbs.append( { "name": pi["name"], "object_id": dir_swhid.object_id.hex(), "path": dir_swhid.path.decode("utf-8") if dir_swhid.path else "", "url": reverse( "swhid-iframe", url_args={"swhid": str(dir_swhid)}, query_params={"focus_swhid": focus_swhid}, ), } ) if filename: breadcrumbs.append( { "name": filename, "object_id": swhid.object_id.hex(), "path": path, "url": "", } ) return breadcrumbs @xframe_options_exempt def swhid_iframe(request, swhid: str): """Django view that can be embedded in an iframe to display objects archived by Software Heritage (currently contents and directories) in a minimalist Web UI. """ focus_swhid = request.GET.get("focus_swhid", swhid) parsed_swhid = None view_data = {} breadcrumbs = [] swh_objects = [] snapshot_context = None swhids_info_extra_context = {} try: parsed_swhid = get_swhid(swhid) parsed_focus_swhid = get_swhid(focus_swhid) path = parsed_swhid.path.decode("utf-8") if parsed_swhid.path else "" snapshot_context = None revision_id = None if ( parsed_swhid.anchor and parsed_swhid.anchor.object_type == ObjectType.REVISION ): revision_id = parsed_swhid.anchor.object_id.hex() if parsed_swhid.origin or parsed_swhid.visit: snapshot_context = get_snapshot_context( origin_url=parsed_swhid.origin, snapshot_id=parsed_swhid.visit.object_id.hex() if parsed_swhid.visit else None, revision_id=revision_id, ) error_info: Dict[str, Any] = {"status_code": 200, "description": ""} if parsed_swhid and parsed_swhid.object_type == ObjectType.CONTENT: view_data = _get_content_rendering_data(parsed_swhid, path) swh_objects.append( SWHObjectInfo( object_type=ObjectType.CONTENT, object_id=parsed_swhid.object_id.hex(), ) ) elif parsed_swhid and parsed_swhid.object_type == ObjectType.DIRECTORY: view_data = _get_directory_rendering_data( parsed_swhid, parsed_focus_swhid, path ) swh_objects.append( SWHObjectInfo( object_type=ObjectType.DIRECTORY, object_id=parsed_swhid.object_id.hex(), ) ) elif parsed_swhid: error_info = { "status_code": 400, "description": ( f"Objects of type {parsed_swhid.object_type} are not supported" ), } swhids_info_extra_context["path"] = path if parsed_swhid and view_data: breadcrumbs = _get_breacrumbs_data( parsed_swhid, parsed_focus_swhid, path, snapshot_context ) if parsed_swhid.object_type == ObjectType.CONTENT and len(breadcrumbs) > 1: swh_objects.append( SWHObjectInfo( object_type=ObjectType.DIRECTORY, object_id=breadcrumbs[-2]["object_id"], ) ) swhids_info_extra_context["path"] = breadcrumbs[-2]["path"] swhids_info_extra_context["filename"] = breadcrumbs[-1]["name"] if snapshot_context: swh_objects.append( SWHObjectInfo( object_type=ObjectType.REVISION, object_id=snapshot_context["revision_id"] or "", ) ) swh_objects.append( SWHObjectInfo( object_type=ObjectType.SNAPSHOT, object_id=snapshot_context["snapshot_id"] or "", ) ) except BadInputExc as e: error_info = {"status_code": 400, "description": f"BadInputExc: {str(e)}"} except NotFoundExc as e: error_info = {"status_code": 404, "description": f"NotFoundExc: {str(e)}"} except Exception as e: error_info = {"status_code": 500, "description": str(e)} return render( request, "misc/iframe.html", { **view_data, "iframe_mode": True, "object_type": parsed_swhid.object_type.value if parsed_swhid else None, "lines": parsed_swhid.lines if parsed_swhid else None, "breadcrumbs": breadcrumbs, "swhid": swhid, "focus_swhid": focus_swhid, "error_code": error_info["status_code"], "error_message": http_status_code_message.get(error_info["status_code"]), "error_description": error_info["description"], "snapshot_context": None, "swhids_info": get_swhids_info( swh_objects, snapshot_context, swhids_info_extra_context ), }, status=error_info["status_code"], ) urlpatterns = [ url( r"^embed/(?Pswh:[0-9]+:[a-z]+:[0-9a-f]+.*)$", swhid_iframe, name="swhid-iframe", ), ] diff --git a/swh/web/templates/browse/help.html b/swh/web/templates/browse/help.html index 803ecc32..60f72a5d 100644 --- a/swh/web/templates/browse/help.html +++ b/swh/web/templates/browse/help.html @@ -1,189 +1,189 @@ {% extends "./layout.html" %} {% comment %} 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 {% endcomment %} {% block navbar-content %}

How to browse the archive ?

{% endblock %} {% block browse-content %}

Overview

This web application aims to provide HTML views to easily navigate in the Software Heritage archive. This is an ongoing development and new features and improvements will be progressively added over time.

URI scheme

The current URI scheme of that web application is described below and depends on the type of Software Heritage object to browse. Its exhaustive documentation can be consulted from the official Software Heritage development documentation

Context-independent browsing

Context-independent URLs provide information about objects (e.g., revisions, directories, contents, persons, …), independently of the contexts where they have been found (e.g., specific software origins, branches, commits, …).

Below are some examples of endpoints used to just render the corresponding information for user consumption:

Where hyperlinks are created when browsing these kind of endpoints, they always point to other context-independent browsing URLs.

Context-dependent browsing

Context-dependent URLs provide information about objects, limited to specific contexts where the objects have been found.

Currently, browsing the Software Heritage objects in the context of an origin is available. Below are some examples of such endpoints:

Search software origins to browse

In order to facilitate the browsing of the archive and generate relevant entry points to it, a search interface is available. Currently, it enables to search software origins from the URLs they were retrieved from. More search criteria will be added in the future. {% endblock %} diff --git a/swh/web/tests/api/test_apidoc.py b/swh/web/tests/api/test_apidoc.py index 5c6e27d7..b206380c 100644 --- a/swh/web/tests/api/test_apidoc.py +++ b/swh/web/tests/api/test_apidoc.py @@ -1,488 +1,488 @@ # Copyright (C) 2015-2019 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 textwrap import pytest from rest_framework.response import Response from swh.storage.exc import StorageAPIError, StorageDBError from swh.web.api.apidoc import _parse_httpdomain_doc, api_doc from swh.web.api.apiurls import api_route from swh.web.common.exc import BadInputExc, ForbiddenExc, NotFoundExc from swh.web.common.utils import prettify_html, reverse from swh.web.tests.utils import check_api_get_responses, check_html_get_response _httpdomain_doc = """ .. http:get:: /api/1/revision/(sha1_git)/ Get information about a revision in the archive. Revisions are identified by **sha1** checksums, compatible with Git commit identifiers. - See :func:`swh.model.identifiers.revision_identifier` in our data model + See :func:`swh.model.git_objects.revision_git_object` in our data model module for details about how they are computed. :param string sha1_git: hexadecimal representation of the revision **sha1_git** identifier :reqheader Accept: the requested response content type, either ``application/json`` (default) or ``application/yaml`` :resheader Content-Type: this depends on :http:header:`Accept` header of request :json object author: information about the author of the revision :>json object committer: information about the committer of the revision :>json string committer_date: ISO representation of the commit date (in UTC) :>json string date: ISO representation of the revision date (in UTC) :>json string directory: the unique identifier that revision points to :>json string directory_url: link to :http:get:`/api/1/directory/(sha1_git)/[(path)/]` to get information about the directory associated to the revision :>json string id: the revision unique identifier :>json boolean merge: whether or not the revision corresponds to a merge commit :>json string message: the message associated to the revision :>json array parents: the parents of the revision, i.e. the previous revisions that head directly to it, each entry of that array contains an unique parent revision identifier but also a link to :http:get:`/api/1/revision/(sha1_git)/` to get more information about it :>json string type: the type of the revision :statuscode 200: no error :statuscode 400: an invalid **sha1_git** value has been provided :statuscode 404: requested revision can not be found in the archive **Example:** .. parsed-literal:: :swh_web_api:`revision/aafb16d69fd30ff58afdd69036a26047f3aebdc6/` """ _exception_http_code = { BadInputExc: 400, ForbiddenExc: 403, NotFoundExc: 404, Exception: 500, StorageAPIError: 503, StorageDBError: 503, } def test_apidoc_nodoc_failure(): with pytest.raises(Exception): @api_doc("/my/nodoc/url/") def apidoc_nodoc_tester(request, arga=0, argb=0): return Response(arga + argb) @api_route(r"/some/(?P[0-9]+)/(?P[0-9]+)/", "api-1-some-doc-route") @api_doc("/some/doc/route/") def apidoc_route(request, myarg, myotherarg, akw=0): """ Sample doc """ return {"result": int(myarg) + int(myotherarg) + akw} def test_apidoc_route_doc(client): url = reverse("api-1-some-doc-route-doc") check_html_get_response( client, url, status_code=200, template_used="api/apidoc.html" ) def test_apidoc_route_fn(api_client): url = reverse("api-1-some-doc-route", url_args={"myarg": 1, "myotherarg": 1}) check_api_get_responses(api_client, url, status_code=200) @api_route(r"/test/error/(?P.+)/", "api-1-test-error") @api_doc("/test/error/") def apidoc_test_error_route(request, exc_name): """ Sample doc """ for e in _exception_http_code.keys(): if e.__name__ == exc_name: raise e("Error") def test_apidoc_error(api_client): for exc, code in _exception_http_code.items(): url = reverse("api-1-test-error", url_args={"exc_name": exc.__name__}) check_api_get_responses(api_client, url, status_code=code) @api_route( r"/some/full/(?P[0-9]+)/(?P[0-9]+)/", "api-1-some-complete-doc-route", ) @api_doc("/some/complete/doc/route/") def apidoc_full_stack(request, myarg, myotherarg, akw=0): """ Sample doc """ return {"result": int(myarg) + int(myotherarg) + akw} def test_apidoc_full_stack_doc(client): url = reverse("api-1-some-complete-doc-route-doc") check_html_get_response( client, url, status_code=200, template_used="api/apidoc.html" ) def test_apidoc_full_stack_fn(api_client): url = reverse( "api-1-some-complete-doc-route", url_args={"myarg": 1, "myotherarg": 1} ) check_api_get_responses(api_client, url, status_code=200) @api_route(r"/test/post/only/", "api-1-test-post-only", methods=["POST"]) @api_doc("/test/post/only/") def apidoc_test_post_only(request, exc_name): """ Sample doc """ return {"result": "some data"} def test_apidoc_post_only(client): # a dedicated view accepting GET requests should have # been created to display the HTML documentation url = reverse("api-1-test-post-only-doc") check_html_get_response( client, url, status_code=200, template_used="api/apidoc.html" ) def test_api_doc_parse_httpdomain(): doc_data = { "description": "", "urls": [], "args": [], "params": [], "resheaders": [], "reqheaders": [], "input_type": "", "inputs": [], "return_type": "", "returns": [], "status_codes": [], "examples": [], } _parse_httpdomain_doc(_httpdomain_doc, doc_data) expected_urls = [ { "rule": "/api/1/revision/ **\\(sha1_git\\)** /", "methods": ["GET", "HEAD", "OPTIONS"], } ] assert "urls" in doc_data assert doc_data["urls"] == expected_urls expected_description = ( "Get information about a revision in the archive. " "Revisions are identified by **sha1** checksums, " "compatible with Git commit identifiers. See " - "**swh.model.identifiers.revision_identifier** in " + "**swh.model.git_objects.revision_git_object** in " "our data model module for details about how they " "are computed." ) assert "description" in doc_data assert doc_data["description"] == expected_description expected_args = [ { "name": "sha1_git", "type": "string", "doc": ( "hexadecimal representation of the revision " "**sha1_git** identifier" ), } ] assert "args" in doc_data assert doc_data["args"] == expected_args expected_params = [] assert "params" in doc_data assert doc_data["params"] == expected_params expected_reqheaders = [ { "doc": ( "the requested response content type, either " "``application/json`` (default) or ``application/yaml``" ), "name": "Accept", } ] assert "reqheaders" in doc_data assert doc_data["reqheaders"] == expected_reqheaders expected_resheaders = [ {"doc": "this depends on **Accept** header of request", "name": "Content-Type"} ] assert "resheaders" in doc_data assert doc_data["resheaders"] == expected_resheaders expected_statuscodes = [ {"code": "200", "doc": "no error"}, {"code": "400", "doc": "an invalid **sha1_git** value has been provided"}, {"code": "404", "doc": "requested revision can not be found in the archive"}, ] assert "status_codes" in doc_data assert doc_data["status_codes"] == expected_statuscodes expected_input_type = "object" assert "input_type" in doc_data assert doc_data["input_type"] == expected_input_type expected_inputs = [ {"name": "n", "type": "int", "doc": "sample input integer"}, {"name": "s", "type": "string", "doc": "sample input string"}, {"name": "a", "type": "array", "doc": "sample input array"}, ] assert "inputs" in doc_data assert doc_data["inputs"] == expected_inputs expected_return_type = "object" assert "return_type" in doc_data assert doc_data["return_type"] == expected_return_type expected_returns = [ { "name": "author", "type": "object", "doc": "information about the author of the revision", }, { "name": "committer", "type": "object", "doc": "information about the committer of the revision", }, { "name": "committer_date", "type": "string", "doc": "ISO representation of the commit date (in UTC)", }, { "name": "date", "type": "string", "doc": "ISO representation of the revision date (in UTC)", }, { "name": "directory", "type": "string", "doc": "the unique identifier that revision points to", }, { "name": "directory_url", "type": "string", "doc": ( "link to `/api/1/directory/ `_ " "to get information about the directory associated to " "the revision" ), }, {"name": "id", "type": "string", "doc": "the revision unique identifier"}, { "name": "merge", "type": "boolean", "doc": "whether or not the revision corresponds to a merge commit", }, { "name": "message", "type": "string", "doc": "the message associated to the revision", }, { "name": "parents", "type": "array", "doc": ( "the parents of the revision, i.e. the previous revisions " "that head directly to it, each entry of that array " "contains an unique parent revision identifier but also a " "link to `/api/1/revision/ `_ " "to get more information about it" ), }, {"name": "type", "type": "string", "doc": "the type of the revision"}, ] assert "returns" in doc_data assert doc_data["returns"] == expected_returns expected_examples = ["/api/1/revision/aafb16d69fd30ff58afdd69036a26047f3aebdc6/"] assert "examples" in doc_data assert doc_data["examples"] == expected_examples @api_route(r"/post/endpoint/", "api-1-post-endpoint", methods=["POST"]) @api_doc("/post/endpoint/") def apidoc_test_post_endpoint(request): """ .. http:post:: /api/1/post/endpoint/ Endpoint documentation :json object : an object whose keys are input SWHIDs and values objects with the following keys: * **known (bool)**: whether the object was found """ pass def test_apidoc_input_output_doc(client): url = reverse("api-1-post-endpoint-doc") rv = check_html_get_response( client, url, status_code=200, template_used="api/apidoc.html" ) input_html_doc = textwrap.indent( ( '
\n' '
\n' " array\n" "
\n" '
\n' "

\n" " Input array of SWHIDs\n" "

\n" "
\n" "
\n" ), " " * 7, ) output_html_doc = textwrap.indent( ( '
\n' '
\n' " object\n" "
\n" '
\n' "

\n" " an object containing the following keys:\n" "

\n" '
\n' "
\n" "
    \n" "
  • \n" "

    \n" " \n" " <swhid> (object)\n" " \n" " : an object whose keys are input SWHIDs" " and values objects with the following keys:\n" "

    \n" "
    \n" '
      \n' "
    • \n" "

      \n" " \n" " known (bool)\n" " \n" " : whether the object was found\n" "

      \n" "
      • \n" "
    \n" "
    \n" "
    • \n" "
\n" "
\n" "
\n" "
\n" "
\n" ), " " * 7, ) html = prettify_html(rv.content) assert input_html_doc in html assert output_html_doc in html @api_route(r"/endpoint/links/in/doc/", "api-1-endpoint-links-in-doc") @api_doc("/endpoint/links/in/doc/") def apidoc_test_endpoint_with_links_in_doc(request): """ .. http:get:: /api/1/post/endpoint/ Endpoint documentation with links to :http:get:`/api/1/content/[(hash_type):](hash)/`, :http:get:`/api/1/directory/(sha1_git)/[(path)/]` and `archive `_. """ pass def test_apidoc_with_links(client): url = reverse("api-1-endpoint-links-in-doc") rv = check_html_get_response( client, url, status_code=200, template_used="api/apidoc.html" ) html = prettify_html(rv.content) first_link = textwrap.indent( ( '\n' " /api/1/content/\n" "" ), " " * 9, ) second_link = textwrap.indent( ( '\n' " /api/1/directory/\n" "" ), " " * 9, ) third_link = textwrap.indent( ( '\n' " archive\n" "" ), " " * 9, ) assert first_link in html assert second_link in html assert third_link in html diff --git a/swh/web/tests/api/test_apiresponse.py b/swh/web/tests/api/test_apiresponse.py index d09b3610..e9d87c44 100644 --- a/swh/web/tests/api/test_apiresponse.py +++ b/swh/web/tests/api/test_apiresponse.py @@ -1,190 +1,190 @@ # Copyright (C) 2015-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import json from corsheaders.middleware import ( ACCESS_CONTROL_ALLOW_HEADERS, ACCESS_CONTROL_ALLOW_METHODS, ACCESS_CONTROL_ALLOW_ORIGIN, ) from hypothesis import given -from swh.model.identifiers import ObjectType +from swh.model.swhids import ObjectType from swh.web.api.apiresponse import ( compute_link_header, filter_by_fields, make_api_response, transform, ) from swh.web.common.identifiers import gen_swhid from swh.web.common.utils import reverse from swh.web.tests.django_asserts import assert_contains from swh.web.tests.strategies import content, directory, revision from swh.web.tests.utils import check_http_get_response, check_http_post_response def test_compute_link_header(): next_link = "/api/endpoint/next" prev_link = "/api/endpoint/prev" rv = { "headers": {"link-next": next_link, "link-prev": prev_link}, "results": [1, 2, 3], } options = {} headers = compute_link_header(rv, options) assert headers == { "Link": (f'<{next_link}>; rel="next",' f'<{prev_link}>; rel="previous"') } def test_compute_link_header_nothing_changed(): rv = {} options = {} headers = compute_link_header(rv, options) assert headers == {} def test_compute_link_header_nothing_changed_2(): rv = {"headers": {}} options = {} headers = compute_link_header(rv, options) assert headers == {} def test_transform_only_return_results_1(): rv = {"results": {"some-key": "some-value"}} assert transform(rv) == {"some-key": "some-value"} def test_transform_only_return_results_2(): rv = {"headers": {"something": "do changes"}, "results": {"some-key": "some-value"}} assert transform(rv) == {"some-key": "some-value"} def test_transform_do_remove_headers(): rv = {"headers": {"something": "do changes"}, "some-key": "some-value"} assert transform(rv) == {"some-key": "some-value"} def test_transform_do_nothing(): rv = {"some-key": "some-value"} assert transform(rv) == {"some-key": "some-value"} def test_swh_multi_response_mimetype(mocker, api_request_factory): mock_shorten_path = mocker.patch("swh.web.api.apiresponse.shorten_path") mock_filter = mocker.patch("swh.web.api.apiresponse.filter_by_fields") mock_json = mocker.patch("swh.web.api.apiresponse.json") data = {"data": [12, 34], "id": "adc83b19e793491b1c6ea0fd8b46cd9f32e592fc"} mock_filter.return_value = data mock_shorten_path.return_value = "my_short_path" mock_json.dumps.return_value = json.dumps(data) accepted_response_formats = { "html": "text/html", "yaml": "application/yaml", "json": "application/json", } for resp_format in accepted_response_formats: request = api_request_factory.get("/api/test/path/") content_type = accepted_response_formats[resp_format] setattr(request, "accepted_media_type", content_type) rv = make_api_response(request, data) mock_filter.assert_called_with(request, data) if resp_format != "html": assert rv.status_code == 200, rv.data assert rv.data == data else: assert rv.status_code == 200, rv.content assert_contains(rv, json.dumps(data)) def test_swh_filter_renderer_do_nothing(api_request_factory): input_data = {"a": "some-data"} request = api_request_factory.get("/api/test/path/", data={}) setattr(request, "query_params", request.GET) actual_data = filter_by_fields(request, input_data) assert actual_data == input_data def test_swh_filter_renderer_do_filter(mocker, api_request_factory): mock_ffk = mocker.patch("swh.web.api.apiresponse.utils.filter_field_keys") mock_ffk.return_value = {"a": "some-data"} request = api_request_factory.get("/api/test/path/", data={"fields": "a,c"}) setattr(request, "query_params", request.GET) input_data = {"a": "some-data", "b": "some-other-data"} actual_data = filter_by_fields(request, input_data) assert actual_data == {"a": "some-data"} mock_ffk.assert_called_once_with(input_data, {"a", "c"}) def test_error_response_handler(mocker, api_client): mock_archive = mocker.patch("swh.web.api.views.stat.archive") mock_archive.stat_counters.side_effect = Exception("Something went wrong") url = reverse("api-1-stat-counters") resp = api_client.get(url) assert resp.status_code == 500 assert "traceback" in resp.data assert "Traceback" in resp.data["traceback"] @given(content(), directory(), revision()) def test_api_endpoints_have_cors_headers(client, content, directory, revision): url = reverse("api-1-stat-counters") resp = check_http_get_response( client, url, status_code=200, http_origin="https://example.org" ) assert ACCESS_CONTROL_ALLOW_ORIGIN in resp swhids = [ gen_swhid(ObjectType.CONTENT, content["sha1_git"]), gen_swhid(ObjectType.DIRECTORY, directory), gen_swhid(ObjectType.REVISION, revision), ] url = reverse("api-1-known") ac_request_method = "POST" ac_request_headers = "Content-Type" resp = client.options( url, HTTP_ORIGIN="https://example.org", HTTP_ACCESS_CONTROL_REQUEST_METHOD=ac_request_method, HTTP_ACCESS_CONTROL_REQUEST_HEADERS=ac_request_headers, ) assert resp.status_code == 200 assert ACCESS_CONTROL_ALLOW_ORIGIN in resp assert ACCESS_CONTROL_ALLOW_METHODS in resp assert ac_request_method in resp[ACCESS_CONTROL_ALLOW_METHODS] assert ACCESS_CONTROL_ALLOW_HEADERS in resp assert ac_request_headers.lower() in resp[ACCESS_CONTROL_ALLOW_HEADERS] resp = resp = check_http_post_response( client, url, data=swhids, status_code=200, http_origin="https://example.org" ) assert ACCESS_CONTROL_ALLOW_ORIGIN in resp diff --git a/swh/web/tests/api/views/test_graph.py b/swh/web/tests/api/views/test_graph.py index b3cb2ba7..63d7a25c 100644 --- a/swh/web/tests/api/views/test_graph.py +++ b/swh/web/tests/api/views/test_graph.py @@ -1,261 +1,261 @@ # Copyright (C) 2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import hashlib import json import textwrap from hypothesis import given from django.http.response import StreamingHttpResponse from swh.model.hashutil import hash_to_bytes -from swh.model.identifiers import ExtendedObjectType, ExtendedSWHID +from swh.model.swhids import ExtendedObjectType, ExtendedSWHID from swh.web.api.views.graph import API_GRAPH_PERM from swh.web.common.utils import reverse from swh.web.config import SWH_WEB_INTERNAL_SERVER_NAME, get_config from swh.web.tests.strategies import origin from swh.web.tests.utils import check_http_get_response def test_graph_endpoint_no_authentication_for_vpn_users(api_client, requests_mock): graph_query = "stats" url = reverse("api-1-graph", url_args={"graph_query": graph_query}) requests_mock.get( get_config()["graph"]["server_url"] + graph_query, json={}, headers={"Content-Type": "application/json"}, ) check_http_get_response( api_client, url, status_code=200, server_name=SWH_WEB_INTERNAL_SERVER_NAME ) def test_graph_endpoint_needs_authentication(api_client): url = reverse("api-1-graph", url_args={"graph_query": "stats"}) check_http_get_response(api_client, url, status_code=401) def _authenticate_graph_user(api_client, keycloak_oidc): keycloak_oidc.client_permissions = [API_GRAPH_PERM] oidc_profile = keycloak_oidc.login() api_client.credentials(HTTP_AUTHORIZATION=f"Bearer {oidc_profile['refresh_token']}") def test_graph_endpoint_needs_permission(api_client, keycloak_oidc, requests_mock): graph_query = "stats" url = reverse("api-1-graph", url_args={"graph_query": graph_query}) oidc_profile = keycloak_oidc.login() api_client.credentials(HTTP_AUTHORIZATION=f"Bearer {oidc_profile['refresh_token']}") check_http_get_response(api_client, url, status_code=403) _authenticate_graph_user(api_client, keycloak_oidc) requests_mock.get( get_config()["graph"]["server_url"] + graph_query, json={}, headers={"Content-Type": "application/json"}, ) check_http_get_response(api_client, url, status_code=200) def test_graph_text_plain_response(api_client, keycloak_oidc, requests_mock): _authenticate_graph_user(api_client, keycloak_oidc) graph_query = "leaves/swh:1:dir:432d1b21c1256f7408a07c577b6974bbdbcc1323" response_text = textwrap.dedent( """\ swh:1:cnt:1d3dace0a825b0535c37c53ed669ef817e9c1b47 swh:1:cnt:6d5b280f4e33589ae967a7912a587dd5cb8dedaa swh:1:cnt:91bef238bf01356a550d416d14bb464c576ac6f4 swh:1:cnt:58a8b925a463b87d49639fda282b8f836546e396 swh:1:cnt:fd32ee0a87e16ccc853dfbeb7018674f9ce008c0 swh:1:cnt:ab7c39871872589a4fc9e249ebc927fb1042c90d swh:1:cnt:93073c02bf3869845977527de16af4d54765838d swh:1:cnt:4251f795b52c54c447a97c9fe904d8b1f993b1e0 swh:1:cnt:c6e7055424332006d07876ffeba684e7e284b383 swh:1:cnt:8459d8867dc3b15ef7ae9683e21cccc9ab2ec887 swh:1:cnt:5f9981d52202815aa947f85b9dfa191b66f51138 swh:1:cnt:00a685ec51bcdf398c15d588ecdedb611dbbab4b swh:1:cnt:e1cf1ea335106a0197a2f92f7804046425a7d3eb swh:1:cnt:07069b38087f88ec192d2c9aff75a502476fd17d swh:1:cnt:f045ee845c7f14d903a2c035b2691a7c400c01f0 """ ) requests_mock.get( get_config()["graph"]["server_url"] + graph_query, text=response_text, headers={"Content-Type": "text/plain", "Transfer-Encoding": "chunked"}, ) url = reverse("api-1-graph", url_args={"graph_query": graph_query}) resp = check_http_get_response( api_client, url, status_code=200, content_type="text/plain" ) assert isinstance(resp, StreamingHttpResponse) assert b"".join(resp.streaming_content) == response_text.encode() _response_json = { "counts": {"nodes": 17075708289, "edges": 196236587976}, "ratios": { "compression": 0.16, "bits_per_node": 58.828, "bits_per_edge": 5.119, "avg_locality": 2184278529.729, }, "indegree": {"min": 0, "max": 263180117, "avg": 11.4921492364925}, "outdegree": {"min": 0, "max": 1033207, "avg": 11.4921492364925}, } def test_graph_json_response(api_client, keycloak_oidc, requests_mock): _authenticate_graph_user(api_client, keycloak_oidc) graph_query = "stats" requests_mock.get( get_config()["graph"]["server_url"] + graph_query, json=_response_json, headers={"Content-Type": "application/json"}, ) url = reverse("api-1-graph", url_args={"graph_query": graph_query}) resp = check_http_get_response(api_client, url, status_code=200) assert resp.content_type == "application/json" assert resp.content == json.dumps(_response_json).encode() def test_graph_ndjson_response(api_client, keycloak_oidc, requests_mock): _authenticate_graph_user(api_client, keycloak_oidc) graph_query = "visit/paths/swh:1:dir:644dd466d8ad527ea3a609bfd588a3244e6dafcb" response_ndjson = textwrap.dedent( """\ ["swh:1:dir:644dd466d8ad527ea3a609bfd588a3244e6dafcb",\ "swh:1:cnt:acfb7cabd63b368a03a9df87670ece1488c8bce0"] ["swh:1:dir:644dd466d8ad527ea3a609bfd588a3244e6dafcb",\ "swh:1:cnt:2a0837708151d76edf28fdbb90dc3eabc676cff3"] ["swh:1:dir:644dd466d8ad527ea3a609bfd588a3244e6dafcb",\ "swh:1:cnt:eaf025ad54b94b2fdda26af75594cfae3491ec75"] """ ) requests_mock.get( get_config()["graph"]["server_url"] + graph_query, text=response_ndjson, headers={ "Content-Type": "application/x-ndjson", "Transfer-Encoding": "chunked", }, ) url = reverse("api-1-graph", url_args={"graph_query": graph_query}) resp = check_http_get_response(api_client, url, status_code=200) assert isinstance(resp, StreamingHttpResponse) assert resp["Content-Type"] == "application/x-ndjson" assert b"".join(resp.streaming_content) == response_ndjson.encode() @given(origin()) def test_graph_response_resolve_origins( archive_data, api_client, keycloak_oidc, requests_mock, origin ): hasher = hashlib.sha1() hasher.update(origin["url"].encode()) origin_sha1 = hasher.digest() origin_swhid = str( ExtendedSWHID(object_type=ExtendedObjectType.ORIGIN, object_id=origin_sha1) ) snapshot = archive_data.snapshot_get_latest(origin["url"])["id"] snapshot_swhid = str( ExtendedSWHID( object_type=ExtendedObjectType.SNAPSHOT, object_id=hash_to_bytes(snapshot) ) ) _authenticate_graph_user(api_client, keycloak_oidc) for graph_query, response_text, content_type in ( ( f"visit/nodes/{snapshot_swhid}", f"{snapshot_swhid}\n{origin_swhid}\n", "text/plain", ), ( f"visit/edges/{snapshot_swhid}", f"{snapshot_swhid} {origin_swhid}\n", "text/plain", ), ( f"visit/paths/{snapshot_swhid}", f'["{snapshot_swhid}", "{origin_swhid}"]\n', "application/x-ndjson", ), ): # set two lines response to check resolved origins cache response_text = response_text + response_text requests_mock.get( get_config()["graph"]["server_url"] + graph_query, text=response_text, headers={"Content-Type": content_type, "Transfer-Encoding": "chunked"}, ) url = reverse( "api-1-graph", url_args={"graph_query": graph_query}, query_params={"direction": "backward"}, ) resp = check_http_get_response(api_client, url, status_code=200) assert isinstance(resp, StreamingHttpResponse) assert resp["Content-Type"] == content_type assert b"".join(resp.streaming_content) == response_text.encode() url = reverse( "api-1-graph", url_args={"graph_query": graph_query}, query_params={"direction": "backward", "resolve_origins": "true"}, ) resp = check_http_get_response(api_client, url, status_code=200) assert isinstance(resp, StreamingHttpResponse) assert resp["Content-Type"] == content_type assert ( b"".join(resp.streaming_content) == response_text.replace(origin_swhid, origin["url"]).encode() ) def test_graph_response_resolve_origins_nothing_to_do( api_client, keycloak_oidc, requests_mock ): _authenticate_graph_user(api_client, keycloak_oidc) graph_query = "stats" requests_mock.get( get_config()["graph"]["server_url"] + graph_query, json=_response_json, headers={"Content-Type": "application/json"}, ) url = reverse( "api-1-graph", url_args={"graph_query": graph_query}, query_params={"resolve_origins": "true"}, ) resp = check_http_get_response(api_client, url, status_code=200) assert resp.content_type == "application/json" assert resp.content == json.dumps(_response_json).encode() diff --git a/swh/web/tests/api/views/test_identifiers.py b/swh/web/tests/api/views/test_identifiers.py index 9e672fd8..c65b8bf0 100644 --- a/swh/web/tests/api/views/test_identifiers.py +++ b/swh/web/tests/api/views/test_identifiers.py @@ -1,188 +1,188 @@ # Copyright (C) 2018-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from hypothesis import given -from swh.model.identifiers import ObjectType +from swh.model.swhids import ObjectType from swh.web.common.identifiers import gen_swhid from swh.web.common.utils import reverse from swh.web.tests.data import random_sha1 from swh.web.tests.strategies import ( content, directory, origin, release, revision, snapshot, unknown_content, unknown_directory, unknown_release, unknown_revision, unknown_snapshot, ) from swh.web.tests.utils import check_api_get_responses, check_api_post_responses @given(origin(), content(), directory(), release(), revision(), snapshot()) def test_swhid_resolve_success( api_client, client, origin, content, directory, release, revision, snapshot ): for obj_type, obj_id in ( (ObjectType.CONTENT, content["sha1_git"]), (ObjectType.DIRECTORY, directory), (ObjectType.RELEASE, release), (ObjectType.REVISION, revision), (ObjectType.SNAPSHOT, snapshot), ): swhid = gen_swhid(obj_type, obj_id, metadata={"origin": origin["url"]}) url = reverse("api-1-resolve-swhid", url_args={"swhid": swhid}) resp = check_api_get_responses(api_client, url, status_code=200) if obj_type == ObjectType.CONTENT: url_args = {"query_string": "sha1_git:%s" % obj_id} elif obj_type == ObjectType.SNAPSHOT: url_args = {"snapshot_id": obj_id} else: url_args = {"sha1_git": obj_id} obj_type_str = obj_type.name.lower() browse_rev_url = reverse( f"browse-{obj_type_str}", url_args=url_args, query_params={"origin_url": origin["url"]}, request=resp.wsgi_request, ) expected_result = { "browse_url": browse_rev_url, "metadata": {"origin": origin["url"]}, "namespace": "swh", "object_id": obj_id, "object_type": obj_type_str, "scheme_version": 1, } assert resp.data == expected_result def test_swhid_resolve_invalid(api_client): rev_id_invalid = "96db9023b8_foo_50d6c108e9a3" swhid = "swh:1:rev:%s" % rev_id_invalid url = reverse("api-1-resolve-swhid", url_args={"swhid": swhid}) check_api_get_responses(api_client, url, status_code=400) @given( unknown_content(), unknown_directory(), unknown_release(), unknown_revision(), unknown_snapshot(), ) def test_swhid_resolve_not_found( api_client, unknown_content, unknown_directory, unknown_release, unknown_revision, unknown_snapshot, ): for obj_type, obj_id in ( (ObjectType.CONTENT, unknown_content["sha1_git"]), (ObjectType.DIRECTORY, unknown_directory), (ObjectType.RELEASE, unknown_release), (ObjectType.REVISION, unknown_revision), (ObjectType.SNAPSHOT, unknown_snapshot), ): swhid = gen_swhid(obj_type, obj_id) url = reverse("api-1-resolve-swhid", url_args={"swhid": swhid}) check_api_get_responses(api_client, url, status_code=404) def test_swh_origin_id_not_resolvable(api_client): ori_swhid = "swh:1:ori:8068d0075010b590762c6cb5682ed53cb3c13deb" url = reverse("api-1-resolve-swhid", url_args={"swhid": ori_swhid}) check_api_get_responses(api_client, url, status_code=400) @given(content(), directory(), release(), revision(), snapshot()) def test_api_known_swhid_all_present( api_client, content, directory, release, revision, snapshot ): input_swhids = [ gen_swhid(ObjectType.CONTENT, content["sha1_git"]), gen_swhid(ObjectType.DIRECTORY, directory), gen_swhid(ObjectType.REVISION, revision), gen_swhid(ObjectType.RELEASE, release), gen_swhid(ObjectType.SNAPSHOT, snapshot), ] url = reverse("api-1-known") resp = check_api_post_responses(api_client, url, data=input_swhids, status_code=200) assert resp.data == {swhid: {"known": True} for swhid in input_swhids} @given(content(), directory()) def test_api_known_swhid_some_present(api_client, content, directory): content_ = gen_swhid(ObjectType.CONTENT, content["sha1_git"]) directory_ = gen_swhid(ObjectType.DIRECTORY, directory) unknown_revision_ = gen_swhid(ObjectType.REVISION, random_sha1()) unknown_release_ = gen_swhid(ObjectType.RELEASE, random_sha1()) unknown_snapshot_ = gen_swhid(ObjectType.SNAPSHOT, random_sha1()) input_swhids = [ content_, directory_, unknown_revision_, unknown_release_, unknown_snapshot_, ] url = reverse("api-1-known") resp = check_api_post_responses(api_client, url, data=input_swhids, status_code=200) assert resp.data == { content_: {"known": True}, directory_: {"known": True}, unknown_revision_: {"known": False}, unknown_release_: {"known": False}, unknown_snapshot_: {"known": False}, } def test_api_known_invalid_swhid(api_client): invalid_swhid_sha1 = ["swh:1:cnt:8068d0075010b590762c6cb5682ed53cb3c13de;"] invalid_swhid_type = ["swh:1:cnn:8068d0075010b590762c6cb5682ed53cb3c13deb"] url = reverse("api-1-known") check_api_post_responses(api_client, url, data=invalid_swhid_sha1, status_code=400) check_api_post_responses(api_client, url, data=invalid_swhid_type, status_code=400) def test_api_known_raises_large_payload_error(api_client): random_swhid = "swh:1:cnt:8068d0075010b590762c6cb5682ed53cb3c13deb" limit = 10000 err_msg = "The maximum number of SWHIDs this endpoint can receive is 1000" swhids = [random_swhid for i in range(limit)] url = reverse("api-1-known") resp = check_api_post_responses(api_client, url, data=swhids, status_code=413) assert resp.data == {"exception": "LargePayloadExc", "reason": err_msg} diff --git a/swh/web/tests/api/views/test_metadata.py b/swh/web/tests/api/views/test_metadata.py index 7fdb83f0..7d83e664 100644 --- a/swh/web/tests/api/views/test_metadata.py +++ b/swh/web/tests/api/views/test_metadata.py @@ -1,208 +1,208 @@ # Copyright (C) 2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import attr from hypothesis import given from hypothesis.strategies import composite, sampled_from, sets import pytest from swh.model.hypothesis_strategies import ( raw_extrinsic_metadata as raw_extrinsic_metadata_orig, ) from swh.model.hypothesis_strategies import sha1_git -from swh.model.identifiers import ExtendedObjectType, ExtendedSWHID, ObjectType +from swh.model.swhids import ExtendedObjectType, ExtendedSWHID, ObjectType from swh.web.common.utils import reverse from swh.web.tests.api.views.utils import scroll_results from swh.web.tests.utils import check_api_get_responses, check_http_get_response # public Web API endpoint for raw extrinsic metadata does not support # extended SWHIDs so we ensure only core ones will be used in test inputs. @composite def raw_extrinsic_metadata(draw): remd = draw(raw_extrinsic_metadata_orig()) remd = attr.evolve( remd, target=ExtendedSWHID( object_type=ExtendedObjectType(draw(sampled_from(ObjectType)).value), object_id=draw(sha1_git()), ), ) return attr.evolve(remd, id=remd.compute_hash()) @given(raw_extrinsic_metadata()) def test_api_raw_extrinsic_metadata(api_client, subtest, metadata): # ensure archive_data fixture will be reset between each hypothesis # example test run @subtest def test_inner(archive_data): archive_data.metadata_authority_add([metadata.authority]) archive_data.metadata_fetcher_add([metadata.fetcher]) archive_data.raw_extrinsic_metadata_add([metadata]) authority = metadata.authority url = reverse( "api-1-raw-extrinsic-metadata-swhid", url_args={"target": str(metadata.target)}, query_params={"authority": f"{authority.type.value} {authority.url}"}, ) rv = check_api_get_responses(api_client, url, status_code=200) assert len(rv.data) == 1 expected_result = metadata.to_dict() del expected_result["id"] del expected_result["metadata"] metadata_url = rv.data[0]["metadata_url"] expected_result["metadata_url"] = metadata_url expected_result["discovery_date"] = expected_result[ "discovery_date" ].isoformat() assert rv.data == [expected_result] rv = check_http_get_response(api_client, metadata_url, status_code=200) assert rv["Content-Type"] == "application/octet-stream" assert ( rv["Content-Disposition"] == f'attachment; filename="{metadata.target}_metadata"' ) assert rv.content == metadata.metadata @pytest.mark.parametrize("limit", [1, 2, 10, 100]) @given(sets(raw_extrinsic_metadata(), min_size=1)) def test_api_raw_extrinsic_metadata_scroll(api_client, subtest, limit, meta): # ensure archive_data fixture will be reset between each hypothesis # example test run @subtest def test_inner(archive_data): # Make all metadata objects use the same authority and target metadata0 = next(iter(meta)) metadata = { attr.evolve(m, authority=metadata0.authority, target=metadata0.target) for m in meta } # Metadata ids must also be updated as they depend on authority and target metadata = {attr.evolve(m, id=m.compute_hash()) for m in metadata} authority = metadata0.authority archive_data.metadata_authority_add([authority]) archive_data.metadata_fetcher_add(list({m.fetcher for m in metadata})) archive_data.raw_extrinsic_metadata_add(metadata) url = reverse( "api-1-raw-extrinsic-metadata-swhid", url_args={"target": str(metadata0.target)}, query_params={ "authority": f"{authority.type.value} {authority.url}", "limit": limit, }, ) results = scroll_results(api_client, url) expected_results = [m.to_dict() for m in metadata] for expected_result in expected_results: del expected_result["id"] del expected_result["metadata"] expected_result["discovery_date"] = expected_result[ "discovery_date" ].isoformat() assert len(results) == len(expected_results) for result in results: del result["metadata_url"] assert result in expected_results _swhid = "swh:1:dir:a2faa28028657859c16ff506924212b33f0e1307" @pytest.mark.parametrize( "status_code,url_args,query_params", [ pytest.param( 200, {"target": _swhid}, {"authority": "forge http://example.org"}, id="minimal working", ), pytest.param( 200, {"target": _swhid}, { "authority": "forge http://example.org", "after": "2021-06-18T09:31:09", "limit": 100, }, id="maximal working", ), pytest.param( 400, {"target": _swhid}, {"authority": "foo http://example.org"}, id="invalid authority type", ), pytest.param( 400, {"target": _swhid}, {"authority": "forge http://example.org", "after": "yesterday",}, id="invalid 'after' format", ), pytest.param( 400, {"target": _swhid}, {"authority": "forge http://example.org", "limit": "abc",}, id="invalid 'limit'", ), ], ) def test_api_raw_extrinsic_metadata_check_params( api_client, archive_data, status_code, url_args, query_params ): url = reverse( "api-1-raw-extrinsic-metadata-swhid", url_args=url_args, query_params=query_params, ) check_api_get_responses(api_client, url, status_code=status_code) @given(raw_extrinsic_metadata()) def test_api_raw_extrinsic_metadata_list_authorities(api_client, subtest, metadata): # ensure archive_data fixture will be reset between each hypothesis # example test run @subtest def test_inner(archive_data): archive_data.metadata_authority_add([metadata.authority]) archive_data.metadata_fetcher_add([metadata.fetcher]) archive_data.raw_extrinsic_metadata_add([metadata]) authority = metadata.authority url = reverse( "api-1-raw-extrinsic-metadata-swhid-authorities", url_args={"target": str(metadata.target)}, ) rv = check_api_get_responses(api_client, url, status_code=200) expected_results = [ { "type": authority.type.value, "url": authority.url, "metadata_list_url": "http://testserver" + reverse( "api-1-raw-extrinsic-metadata-swhid", url_args={"target": str(metadata.target)}, query_params={ "authority": f"{authority.type.value} {authority.url}" }, ), } ] assert rv.data == expected_results diff --git a/swh/web/tests/api/views/test_vault.py b/swh/web/tests/api/views/test_vault.py index 5826fbe1..893b5a74 100644 --- a/swh/web/tests/api/views/test_vault.py +++ b/swh/web/tests/api/views/test_vault.py @@ -1,337 +1,337 @@ # Copyright (C) 2017-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import re from hypothesis import given import pytest -from swh.model.identifiers import CoreSWHID +from swh.model.swhids import CoreSWHID from swh.vault.exc import NotFoundExc from swh.web.common.utils import reverse from swh.web.tests.strategies import ( directory, revision, unknown_directory, unknown_revision, ) from swh.web.tests.utils import ( check_api_get_responses, check_api_post_responses, check_http_get_response, check_http_post_response, ) ##################### # Current API: @given(directory(), revision()) def test_api_vault_cook(api_client, mocker, directory, revision): mock_archive = mocker.patch("swh.web.api.views.vault.archive") for bundle_type, swhid, content_type, in ( ("flat", f"swh:1:dir:{directory}", "application/gzip"), ("gitfast", f"swh:1:rev:{revision}", "application/gzip"), ("git_bare", f"swh:1:rev:{revision}", "application/x-tar"), ): swhid = CoreSWHID.from_string(swhid) fetch_url = reverse( f"api-1-vault-fetch-{bundle_type.replace('_', '-')}", url_args={"swhid": str(swhid)}, ) stub_cook = { "type": bundle_type, "progress_msg": None, "task_id": 1, "task_status": "done", "swhid": swhid, } stub_fetch = b"content" mock_archive.vault_cook.return_value = stub_cook mock_archive.vault_fetch.return_value = stub_fetch email = "test@test.mail" url = reverse( f"api-1-vault-cook-{bundle_type.replace('_', '-')}", url_args={"swhid": str(swhid)}, query_params={"email": email}, ) rv = check_api_post_responses(api_client, url, data=None, status_code=200) assert rv.data == { "fetch_url": rv.wsgi_request.build_absolute_uri(fetch_url), "progress_message": None, "id": 1, "status": "done", "swhid": str(swhid), } mock_archive.vault_cook.assert_called_with(bundle_type, swhid, email) rv = check_http_get_response(api_client, fetch_url, status_code=200) assert rv["Content-Type"] == content_type assert rv.content == stub_fetch mock_archive.vault_fetch.assert_called_with(bundle_type, swhid) @given(directory(), revision(), unknown_directory(), unknown_revision()) def test_api_vault_cook_notfound( api_client, mocker, directory, revision, unknown_directory, unknown_revision ): mock_vault = mocker.patch("swh.web.common.archive.vault") mock_vault.cook.side_effect = NotFoundExc("object not found") mock_vault.fetch.side_effect = NotFoundExc("cooked archive not found") mock_vault.progress.side_effect = NotFoundExc("cooking request not found") for bundle_type, swhid in ( ("flat", f"swh:1:dir:{directory}"), ("gitfast", f"swh:1:rev:{revision}"), ("git_bare", f"swh:1:rev:{revision}"), ): swhid = CoreSWHID.from_string(swhid) url = reverse( f"api-1-vault-cook-{bundle_type.replace('_', '-')}", url_args={"swhid": str(swhid)}, ) rv = check_api_get_responses(api_client, url, status_code=404) assert rv.data["exception"] == "NotFoundExc" assert rv.data["reason"] == f"Cooking of {swhid} was never requested." mock_vault.progress.assert_called_with(bundle_type, swhid) for bundle_type, swhid in ( ("flat", f"swh:1:dir:{unknown_directory}"), ("gitfast", f"swh:1:rev:{unknown_revision}"), ("git_bare", f"swh:1:rev:{unknown_revision}"), ): swhid = CoreSWHID.from_string(swhid) url = reverse( f"api-1-vault-cook-{bundle_type.replace('_', '-')}", url_args={"swhid": str(swhid)}, ) rv = check_api_post_responses(api_client, url, data=None, status_code=404) assert rv.data["exception"] == "NotFoundExc" assert rv.data["reason"] == f"{swhid} not found." mock_vault.cook.assert_called_with(bundle_type, swhid, email=None) fetch_url = reverse( f"api-1-vault-fetch-{bundle_type.replace('_', '-')}", url_args={"swhid": str(swhid)}, ) rv = check_api_get_responses(api_client, fetch_url, status_code=404) assert rv.data["exception"] == "NotFoundExc" assert rv.data["reason"] == f"Cooked archive for {swhid} not found." mock_vault.fetch.assert_called_with(bundle_type, swhid) @pytest.mark.parametrize("bundle_type", ["flat", "gitfast", "git_bare"]) def test_api_vault_cook_error_content(api_client, mocker, bundle_type): swhid = "swh:1:cnt:" + "0" * 40 email = "test@test.mail" url = reverse( f"api-1-vault-cook-{bundle_type.replace('_', '-')}", url_args={"swhid": swhid}, query_params={"email": email}, ) rv = check_api_post_responses(api_client, url, data=None, status_code=400) assert rv.data == { "exception": "BadInputExc", "reason": ( "Content objects do not need to be cooked, " "use `/api/1/content/raw/` instead." ), } @pytest.mark.parametrize( "bundle_type,swhid_type,hint", [ ("flat", "rev", True), ("flat", "rel", False), ("flat", "snp", False), ("gitfast", "dir", True), ("gitfast", "rel", False), ("gitfast", "snp", False), ("git_bare", "dir", True), ("git_bare", "rel", False), ("git_bare", "snp", False), ], ) def test_api_vault_cook_error(api_client, mocker, bundle_type, swhid_type, hint): swhid = f"swh:1:{swhid_type}:" + "0" * 40 email = "test@test.mail" url = reverse( f"api-1-vault-cook-{bundle_type.replace('_', '-')}", url_args={"swhid": swhid}, query_params={"email": email}, ) rv = check_api_post_responses(api_client, url, data=None, status_code=400) assert rv.data["exception"] == "BadInputExc" if hint: assert re.match( r"Only .* can be cooked as .* bundles\. Use .*", rv.data["reason"] ) else: assert re.match(r"Only .* can be cooked as .* bundles\.", rv.data["reason"]) ##################### # Legacy API: @given(directory(), revision()) def test_api_vault_cook_legacy(api_client, mocker, directory, revision): mock_archive = mocker.patch("swh.web.api.views.vault.archive") for obj_type, bundle_type, response_obj_type, obj_id in ( ("directory", "flat", "directory", directory), ("revision_gitfast", "gitfast", "revision", revision), ): swhid = CoreSWHID.from_string(f"swh:1:{obj_type[:3]}:{obj_id}") fetch_url = reverse( f"api-1-vault-fetch-{bundle_type}", url_args={"swhid": str(swhid)}, ) stub_cook = { "type": obj_type, "progress_msg": None, "task_id": 1, "task_status": "done", "swhid": swhid, "obj_type": response_obj_type, "obj_id": obj_id, } stub_fetch = b"content" mock_archive.vault_cook.return_value = stub_cook mock_archive.vault_fetch.return_value = stub_fetch email = "test@test.mail" url = reverse( f"api-1-vault-cook-{obj_type}", url_args={f"{obj_type[:3]}_id": obj_id}, query_params={"email": email}, ) rv = check_api_post_responses(api_client, url, data=None, status_code=200) assert rv.data == { "fetch_url": rv.wsgi_request.build_absolute_uri(fetch_url), "progress_message": None, "id": 1, "status": "done", "swhid": str(swhid), "obj_type": response_obj_type, "obj_id": obj_id, } mock_archive.vault_cook.assert_called_with(bundle_type, swhid, email) rv = check_http_get_response(api_client, fetch_url, status_code=200) assert rv["Content-Type"] == "application/gzip" assert rv.content == stub_fetch mock_archive.vault_fetch.assert_called_with(bundle_type, swhid) @given(directory(), revision()) def test_api_vault_cook_uppercase_hash_legacy(api_client, directory, revision): for obj_type, obj_id in ( ("directory", directory), ("revision_gitfast", revision), ): url = reverse( f"api-1-vault-cook-{obj_type}-uppercase-checksum", url_args={f"{obj_type[:3]}_id": obj_id.upper()}, ) rv = check_http_post_response( api_client, url, data={"email": "test@test.mail"}, status_code=302 ) redirect_url = reverse( f"api-1-vault-cook-{obj_type}", url_args={f"{obj_type[:3]}_id": obj_id} ) assert rv["location"] == redirect_url fetch_url = reverse( f"api-1-vault-fetch-{obj_type}-uppercase-checksum", url_args={f"{obj_type[:3]}_id": obj_id.upper()}, ) rv = check_http_get_response(api_client, fetch_url, status_code=302) redirect_url = reverse( f"api-1-vault-fetch-{obj_type}", url_args={f"{obj_type[:3]}_id": obj_id}, ) assert rv["location"] == redirect_url @given(directory(), revision(), unknown_directory(), unknown_revision()) def test_api_vault_cook_notfound_legacy( api_client, mocker, directory, revision, unknown_directory, unknown_revision ): mock_vault = mocker.patch("swh.web.common.archive.vault") mock_vault.cook.side_effect = NotFoundExc("object not found") mock_vault.fetch.side_effect = NotFoundExc("cooked archive not found") mock_vault.progress.side_effect = NotFoundExc("cooking request not found") for obj_type, bundle_type, obj_id in ( ("directory", "flat", directory), ("revision_gitfast", "gitfast", revision), ): url = reverse( f"api-1-vault-cook-{obj_type}", url_args={f"{obj_type[:3]}_id": obj_id}, ) swhid = CoreSWHID.from_string(f"swh:1:{obj_type[:3]}:{obj_id}") rv = check_api_get_responses(api_client, url, status_code=404) assert rv.data["exception"] == "NotFoundExc" assert rv.data["reason"] == f"Cooking of {swhid} was never requested." mock_vault.progress.assert_called_with(bundle_type, swhid) for obj_type, bundle_type, obj_id in ( ("directory", "flat", unknown_directory), ("revision_gitfast", "gitfast", unknown_revision), ): swhid = CoreSWHID.from_string(f"swh:1:{obj_type[:3]}:{obj_id}") url = reverse( f"api-1-vault-cook-{obj_type}", url_args={f"{obj_type[:3]}_id": obj_id} ) rv = check_api_post_responses(api_client, url, data=None, status_code=404) assert rv.data["exception"] == "NotFoundExc" assert rv.data["reason"] == f"{swhid} not found." mock_vault.cook.assert_called_with(bundle_type, swhid, email=None) fetch_url = reverse( f"api-1-vault-fetch-{obj_type}", url_args={f"{obj_type[:3]}_id": obj_id}, ) # Redirected to the current 'fetch' url rv = check_http_get_response(api_client, fetch_url, status_code=302) redirect_url = reverse( f"api-1-vault-fetch-{bundle_type}", url_args={"swhid": str(swhid)}, ) assert rv["location"] == redirect_url rv = check_api_get_responses(api_client, redirect_url, status_code=404) assert rv.data["exception"] == "NotFoundExc" assert rv.data["reason"] == f"Cooked archive for {swhid} not found." mock_vault.fetch.assert_called_with(bundle_type, swhid) diff --git a/swh/web/tests/browse/test_snapshot_context.py b/swh/web/tests/browse/test_snapshot_context.py index c185ffa3..54861cac 100644 --- a/swh/web/tests/browse/test_snapshot_context.py +++ b/swh/web/tests/browse/test_snapshot_context.py @@ -1,412 +1,412 @@ # Copyright (C) 2020-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import random from hypothesis import given -from swh.model.identifiers import ObjectType +from swh.model.swhids import ObjectType from swh.web.browse.snapshot_context import ( _get_release, get_origin_visit_snapshot, get_snapshot_content, get_snapshot_context, ) from swh.web.browse.utils import gen_revision_url from swh.web.common.identifiers import gen_swhid from swh.web.common.origin_visits import get_origin_visit, get_origin_visits from swh.web.common.typing import ( SnapshotBranchInfo, SnapshotContext, SnapshotReleaseInfo, ) from swh.web.common.utils import format_utc_iso_date, reverse from swh.web.tests.strategies import ( origin_with_multiple_visits, origin_with_releases, snapshot, ) @given(origin_with_multiple_visits()) def test_get_origin_visit_snapshot_simple(archive_data, origin): visits = archive_data.origin_visit_get(origin["url"]) for visit in visits: snapshot = archive_data.snapshot_get(visit["snapshot"]) branches = [] releases = [] def _process_branch_data(branch, branch_data, alias=False): if branch_data["target_type"] == "revision": rev_data = archive_data.revision_get(branch_data["target"]) branches.append( SnapshotBranchInfo( name=branch, alias=alias, revision=branch_data["target"], directory=rev_data["directory"], date=format_utc_iso_date(rev_data["date"]), message=rev_data["message"], url=None, ) ) elif branch_data["target_type"] == "release": rel_data = archive_data.release_get(branch_data["target"]) rev_data = archive_data.revision_get(rel_data["target"]) releases.append( SnapshotReleaseInfo( name=rel_data["name"], alias=alias, branch_name=branch, date=format_utc_iso_date(rel_data["date"]), id=rel_data["id"], message=rel_data["message"], target_type=rel_data["target_type"], target=rel_data["target"], directory=rev_data["directory"], url=None, ) ) aliases = {} for branch in sorted(snapshot["branches"].keys()): branch_data = snapshot["branches"][branch] if branch_data["target_type"] == "alias": target_data = snapshot["branches"][branch_data["target"]] aliases[branch] = target_data _process_branch_data(branch, target_data, alias=True) else: _process_branch_data(branch, branch_data) assert branches and releases, "Incomplete test data." origin_visit_branches = get_origin_visit_snapshot( origin, visit_id=visit["visit"] ) assert origin_visit_branches == (branches, releases, aliases) @given(snapshot()) def test_get_snapshot_context_no_origin(archive_data, snapshot): for browse_context, kwargs in ( ("content", {"snapshot_id": snapshot, "path": "/some/path"}), ("directory", {"snapshot_id": snapshot}), ("log", {"snapshot_id": snapshot}), ): url_args = {"snapshot_id": snapshot} query_params = dict(kwargs) query_params.pop("snapshot_id") snapshot_context = get_snapshot_context(**kwargs, browse_context=browse_context) branches, releases, _ = get_snapshot_content(snapshot) releases = list(reversed(releases)) revision_id = None root_directory = None for branch in branches: if branch["name"] == "HEAD": revision_id = branch["revision"] root_directory = branch["directory"] branch["url"] = reverse( f"browse-snapshot-{browse_context}", url_args=url_args, query_params={"branch": branch["name"], **query_params}, ) for release in releases: release["url"] = reverse( f"browse-snapshot-{browse_context}", url_args=url_args, query_params={"release": release["name"], **query_params}, ) branches_url = reverse("browse-snapshot-branches", url_args=url_args) releases_url = reverse("browse-snapshot-releases", url_args=url_args) is_empty = not branches and not releases snapshot_swhid = gen_swhid(ObjectType.SNAPSHOT, snapshot) snapshot_sizes = archive_data.snapshot_count_branches(snapshot) expected = SnapshotContext( branch="HEAD", branch_alias=True, branches=branches, branches_url=branches_url, is_empty=is_empty, origin_info=None, origin_visits_url=None, release=None, release_alias=False, release_id=None, query_params=query_params, releases=releases, releases_url=releases_url, revision_id=revision_id, revision_info=_get_revision_info(archive_data, revision_id), root_directory=root_directory, snapshot_id=snapshot, snapshot_sizes=snapshot_sizes, snapshot_swhid=snapshot_swhid, url_args=url_args, visit_info=None, ) if revision_id: expected["revision_info"]["revision_url"] = gen_revision_url( revision_id, snapshot_context ) assert snapshot_context == expected _check_branch_release_revision_parameters( archive_data, expected, browse_context, kwargs, branches, releases ) @given(origin_with_multiple_visits()) def test_get_snapshot_context_with_origin(archive_data, origin): origin_visits = get_origin_visits(origin) timestamp = format_utc_iso_date(origin_visits[0]["date"], "%Y-%m-%dT%H:%M:%SZ") visit_id = origin_visits[1]["visit"] for browse_context, kwargs in ( ("content", {"origin_url": origin["url"], "path": "/some/path"}), ("directory", {"origin_url": origin["url"]}), ("log", {"origin_url": origin["url"]}), ("directory", {"origin_url": origin["url"], "timestamp": timestamp,},), ("directory", {"origin_url": origin["url"], "visit_id": visit_id,},), ): visit_id = kwargs["visit_id"] if "visit_id" in kwargs else None visit_ts = kwargs["timestamp"] if "timestamp" in kwargs else None visit_info = get_origin_visit( {"url": kwargs["origin_url"]}, visit_ts=visit_ts, visit_id=visit_id ) snapshot = visit_info["snapshot"] snapshot_context = get_snapshot_context(**kwargs, browse_context=browse_context) query_params = dict(kwargs) branches, releases, _ = get_snapshot_content(snapshot) releases = list(reversed(releases)) revision_id = None root_directory = None for branch in branches: if branch["name"] == "HEAD": revision_id = branch["revision"] root_directory = branch["directory"] branch["url"] = reverse( f"browse-origin-{browse_context}", query_params={"branch": branch["name"], **query_params}, ) for release in releases: release["url"] = reverse( f"browse-origin-{browse_context}", query_params={"release": release["name"], **query_params}, ) query_params.pop("path", None) branches_url = reverse("browse-origin-branches", query_params=query_params) releases_url = reverse("browse-origin-releases", query_params=query_params) origin_visits_url = reverse( "browse-origin-visits", query_params={"origin_url": kwargs["origin_url"]} ) is_empty = not branches and not releases snapshot_swhid = gen_swhid(ObjectType.SNAPSHOT, snapshot) snapshot_sizes = archive_data.snapshot_count_branches(snapshot) visit_info["url"] = reverse( "browse-origin-directory", query_params=query_params ) visit_info["formatted_date"] = format_utc_iso_date(visit_info["date"]) if "path" in kwargs: query_params["path"] = kwargs["path"] expected = SnapshotContext( branch="HEAD", branch_alias=True, branches=branches, branches_url=branches_url, is_empty=is_empty, origin_info={"url": origin["url"]}, origin_visits_url=origin_visits_url, release=None, release_alias=False, release_id=None, query_params=query_params, releases=releases, releases_url=releases_url, revision_id=revision_id, revision_info=_get_revision_info(archive_data, revision_id), root_directory=root_directory, snapshot_id=snapshot, snapshot_sizes=snapshot_sizes, snapshot_swhid=snapshot_swhid, url_args={}, visit_info=visit_info, ) if revision_id: expected["revision_info"]["revision_url"] = gen_revision_url( revision_id, snapshot_context ) assert snapshot_context == expected _check_branch_release_revision_parameters( archive_data, expected, browse_context, kwargs, branches, releases ) def _check_branch_release_revision_parameters( archive_data, base_expected_context, browse_context, kwargs, branches, releases, ): branch = random.choice(branches) snapshot_context = get_snapshot_context( **kwargs, browse_context=browse_context, branch_name=branch["name"] ) url_args = dict(kwargs) url_args.pop("path", None) url_args.pop("timestamp", None) url_args.pop("visit_id", None) url_args.pop("origin_url", None) query_params = dict(kwargs) query_params.pop("snapshot_id", None) expected_branch = dict(base_expected_context) expected_branch["branch"] = branch["name"] expected_branch["branch_alias"] = branch["alias"] expected_branch["revision_id"] = branch["revision"] expected_branch["revision_info"] = _get_revision_info( archive_data, branch["revision"] ) expected_branch["root_directory"] = branch["directory"] expected_branch["query_params"] = {"branch": branch["name"], **query_params} expected_branch["revision_info"]["revision_url"] = gen_revision_url( branch["revision"], expected_branch ) assert snapshot_context == expected_branch if releases: release = random.choice(releases) snapshot_context = get_snapshot_context( **kwargs, browse_context=browse_context, release_name=release["name"] ) expected_release = dict(base_expected_context) expected_release["branch"] = None expected_release["branch_alias"] = False expected_release["release"] = release["name"] expected_release["release_id"] = release["id"] if release["target_type"] == "revision": expected_release["revision_id"] = release["target"] expected_release["revision_info"] = _get_revision_info( archive_data, release["target"] ) expected_release["root_directory"] = release["directory"] expected_release["query_params"] = {"release": release["name"], **query_params} expected_release["revision_info"]["revision_url"] = gen_revision_url( release["target"], expected_release ) assert snapshot_context == expected_release revision_log = archive_data.revision_log(branch["revision"]) revision = revision_log[-1] snapshot_context = get_snapshot_context( **kwargs, browse_context=browse_context, revision_id=revision["id"] ) if "origin_url" in kwargs: view_name = f"browse-origin-{browse_context}" else: view_name = f"browse-snapshot-{browse_context}" kwargs.pop("visit_id", None) revision_browse_url = reverse( view_name, url_args=url_args, query_params={"revision": revision["id"], **query_params}, ) branches.append( SnapshotBranchInfo( name=revision["id"], alias=False, revision=revision["id"], directory=revision["directory"], date=revision["date"], message=revision["message"], url=revision_browse_url, ) ) expected_revision = dict(base_expected_context) expected_revision["branch"] = None expected_revision["branch_alias"] = False expected_revision["branches"] = branches expected_revision["revision_id"] = revision["id"] expected_revision["revision_info"] = _get_revision_info( archive_data, revision["id"] ) expected_revision["root_directory"] = revision["directory"] expected_revision["query_params"] = {"revision": revision["id"], **query_params} expected_revision["revision_info"]["revision_url"] = gen_revision_url( revision["id"], expected_revision ) assert snapshot_context == expected_revision @given(origin_with_releases()) def test_get_release_large_snapshot(archive_data, origin): snapshot = archive_data.snapshot_get_latest(origin["url"]) release_id = random.choice( [ v["target"] for v in snapshot["branches"].values() if v["target_type"] == "release" ] ) release_data = archive_data.release_get(release_id) # simulate large snapshot processing by providing releases parameter # as an empty list release = _get_release( releases=[], release_name=release_data["name"], snapshot_id=snapshot["id"] ) assert release_data["name"] == release["name"] assert release_data["id"] == release["id"] def _get_revision_info(archive_data, revision_id): revision_info = None if revision_id: revision_info = archive_data.revision_get(revision_id) revision_info["message_header"] = revision_info["message"].split("\n")[0] revision_info["date"] = format_utc_iso_date(revision_info["date"]) revision_info["committer_date"] = format_utc_iso_date( revision_info["committer_date"] ) return revision_info diff --git a/swh/web/tests/browse/views/test_content.py b/swh/web/tests/browse/views/test_content.py index 186f6763..c33d92e2 100644 --- a/swh/web/tests/browse/views/test_content.py +++ b/swh/web/tests/browse/views/test_content.py @@ -1,619 +1,619 @@ # Copyright (C) 2017-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import random from hypothesis import given from django.utils.html import escape -from swh.model.identifiers import ObjectType +from swh.model.swhids import ObjectType from swh.web.browse.snapshot_context import process_snapshot_branches from swh.web.browse.utils import ( _re_encode_content, get_mimetype_and_encoding_for_content, prepare_content_for_display, ) from swh.web.common.exc import NotFoundExc from swh.web.common.identifiers import gen_swhid from swh.web.common.utils import gen_path_info, reverse from swh.web.tests.django_asserts import assert_contains, assert_not_contains from swh.web.tests.strategies import ( content, content_application_no_highlight, content_image_type, content_text, content_text_no_highlight, content_text_non_utf8, content_unsupported_image_type_rendering, content_utf8_detected_as_binary, invalid_sha1, origin_with_multiple_visits, unknown_content, ) from swh.web.tests.utils import check_html_get_response, check_http_get_response @given(content_text()) def test_content_view_text(client, archive_data, content): sha1_git = content["sha1_git"] url = reverse( "browse-content", url_args={"query_string": content["sha1"]}, query_params={"path": content["path"]}, ) url_raw = reverse("browse-content-raw", url_args={"query_string": content["sha1"]}) resp = check_html_get_response( client, url, status_code=200, template_used="browse/content.html" ) content_display = _process_content_for_display(archive_data, content) mimetype = content_display["mimetype"] if mimetype.startswith("text/"): assert_contains(resp, '' % content_display["language"]) assert_contains(resp, escape(content_display["content_data"])) assert_contains(resp, url_raw) swh_cnt_id = gen_swhid(ObjectType.CONTENT, sha1_git) swh_cnt_id_url = reverse("browse-swhid", url_args={"swhid": swh_cnt_id}) assert_contains(resp, swh_cnt_id) assert_contains(resp, swh_cnt_id_url) assert_not_contains(resp, "swh-metadata-popover") @given(content_application_no_highlight(), content_text_no_highlight()) def test_content_view_no_highlight(client, archive_data, content_app, content_text): for content_ in (content_app, content_text): content = content_ sha1_git = content["sha1_git"] url = reverse("browse-content", url_args={"query_string": content["sha1"]}) url_raw = reverse( "browse-content-raw", url_args={"query_string": content["sha1"]} ) resp = check_html_get_response( client, url, status_code=200, template_used="browse/content.html" ) content_display = _process_content_for_display(archive_data, content) assert_contains(resp, '') assert_contains(resp, escape(content_display["content_data"])) assert_contains(resp, url_raw) swh_cnt_id = gen_swhid(ObjectType.CONTENT, sha1_git) swh_cnt_id_url = reverse("browse-swhid", url_args={"swhid": swh_cnt_id}) assert_contains(resp, swh_cnt_id) assert_contains(resp, swh_cnt_id_url) @given(content_text_non_utf8()) def test_content_view_no_utf8_text(client, archive_data, content): sha1_git = content["sha1_git"] url = reverse("browse-content", url_args={"query_string": content["sha1"]}) resp = check_html_get_response( client, url, status_code=200, template_used="browse/content.html" ) content_display = _process_content_for_display(archive_data, content) swh_cnt_id = gen_swhid(ObjectType.CONTENT, sha1_git) swh_cnt_id_url = reverse("browse-swhid", url_args={"swhid": swh_cnt_id}) assert_contains(resp, swh_cnt_id_url) assert_contains(resp, escape(content_display["content_data"])) @given(content_image_type()) def test_content_view_image(client, archive_data, content): url = reverse("browse-content", url_args={"query_string": content["sha1"]}) url_raw = reverse("browse-content-raw", url_args={"query_string": content["sha1"]}) resp = check_html_get_response( client, url, status_code=200, template_used="browse/content.html" ) content_display = _process_content_for_display(archive_data, content) mimetype = content_display["mimetype"] content_data = content_display["content_data"] assert_contains(resp, '' % (mimetype, content_data)) assert_contains(resp, url_raw) @given(content_unsupported_image_type_rendering()) def test_content_view_image_no_rendering(client, archive_data, content): url = reverse("browse-content", url_args={"query_string": content["sha1"]}) resp = check_html_get_response( client, url, status_code=200, template_used="browse/content.html" ) mimetype = content["mimetype"] encoding = content["encoding"] assert_contains( resp, ( f"Content with mime type {mimetype} and encoding {encoding} " "cannot be displayed." ), ) @given(content_text()) def test_content_view_text_with_path(client, archive_data, content): path = content["path"] url = reverse( "browse-content", url_args={"query_string": content["sha1"]}, query_params={"path": path}, ) resp = check_html_get_response( client, url, status_code=200, template_used="browse/content.html" ) assert_contains(resp, '