diff --git a/swh/web/api/views/content.py b/swh/web/api/views/content.py index 2150d123..5536e883 100644 --- a/swh/web/api/views/content.py +++ b/swh/web/api/views/content.py @@ -1,406 +1,342 @@ # Copyright (C) 2015-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import functools import io from typing import Optional from django.http import FileResponse from rest_framework.request import Request 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 from swh.web.common.exc import NotFoundExc -from swh.web.common.utils import reverse @api_route( r"/content/(?P[0-9a-z_:]*[0-9a-f]+)/filetype/", "api-1-content-filetype", checksum_args=["q"], ) @api_doc("/content/filetype/") @format_docstring() def api_content_filetype(request: Request, q: str): """ .. http:get:: /api/1/content/[(hash_type):](hash)/filetype/ Get information about the detected MIME type of a content object. :param string hash_type: optional parameter specifying which hashing algorithm has been used to compute the content checksum. It can be either ``sha1``, ``sha1_git``, ``sha256`` or ``blake2s256``. If that parameter is not provided, it is assumed that the hashing algorithm used is ``sha1``. :param string hash: hexadecimal representation of the checksum value computed with the specified hashing algorithm. :>json object content_url: link to :http:get:`/api/1/content/[(hash_type):](hash)/` for getting information about the content :>json string encoding: the detected content encoding :>json string id: the **sha1** identifier of the content :>json string mimetype: the detected MIME type of the content :>json object tool: information about the tool used to detect the content filetype {common_headers} :statuscode 200: no error :statuscode 400: an invalid **hash_type** or **hash** has been provided :statuscode 404: requested content can not be found in the archive **Example:** .. parsed-literal:: :swh_web_api:`content/sha1:dc2830a9e72f23c1dfebef4413003221baa5fb62/filetype/` """ return api_lookup( archive.lookup_content_filetype, q, notfound_msg="No filetype information found for content {}.".format(q), enrich_fn=utils.enrich_metadata_endpoint, request=request, ) @api_route( r"/content/(?P[0-9a-z_:]*[0-9a-f]+)/language/", "api-1-content-language", checksum_args=["q"], ) @api_doc("/content/language/") @format_docstring() def api_content_language(request: Request, q: str): """ .. http:get:: /api/1/content/[(hash_type):](hash)/language/ Get information about the programming language used in a content object. Note: this endpoint currently returns no data. :param string hash_type: optional parameter specifying which hashing algorithm has been used to compute the content checksum. It can be either ``sha1``, ``sha1_git``, ``sha256`` or ``blake2s256``. If that parameter is not provided, it is assumed that the hashing algorithm used is ``sha1``. :param string hash: hexadecimal representation of the checksum value computed with the specified hashing algorithm. :>json object content_url: link to :http:get:`/api/1/content/[(hash_type):](hash)/` for getting information about the content :>json string id: the **sha1** identifier of the content :>json string lang: the detected programming language if any :>json object tool: information about the tool used to detect the programming language {common_headers} :statuscode 200: no error :statuscode 400: an invalid **hash_type** or **hash** has been provided :statuscode 404: requested content can not be found in the archive **Example:** .. parsed-literal:: :swh_web_api:`content/sha1:dc2830a9e72f23c1dfebef4413003221baa5fb62/language/` """ return api_lookup( archive.lookup_content_language, q, notfound_msg="No language information found for content {}.".format(q), enrich_fn=utils.enrich_metadata_endpoint, request=request, ) @api_route( r"/content/(?P[0-9a-z_:]*[0-9a-f]+)/license/", "api-1-content-license", checksum_args=["q"], ) @api_doc("/content/license/") @format_docstring() def api_content_license(request: Request, q: str): """ .. http:get:: /api/1/content/[(hash_type):](hash)/license/ Get information about the license of a content object. :param string hash_type: optional parameter specifying which hashing algorithm has been used to compute the content checksum. It can be either ``sha1``, ``sha1_git``, ``sha256`` or ``blake2s256``. If that parameter is not provided, it is assumed that the hashing algorithm used is ``sha1``. :param string hash: hexadecimal representation of the checksum value computed with the specified hashing algorithm. :>json object content_url: link to :http:get:`/api/1/content/[(hash_type):](hash)/` for getting information about the content :>json string id: the **sha1** identifier of the content :>json array licenses: array of strings containing the detected license names :>json object tool: information about the tool used to detect the license {common_headers} :statuscode 200: no error :statuscode 400: an invalid **hash_type** or **hash** has been provided :statuscode 404: requested content can not be found in the archive **Example:** .. parsed-literal:: :swh_web_api:`content/sha1:dc2830a9e72f23c1dfebef4413003221baa5fb62/license/` """ return api_lookup( archive.lookup_content_license, q, notfound_msg="No license information found for content {}.".format(q), enrich_fn=utils.enrich_metadata_endpoint, request=request, ) -@api_route(r"/content/(?P[0-9a-z_:]*[0-9a-f]+)/ctags/", "api-1-content-ctags") -@api_doc("/content/ctags/", tags=["hidden"]) -def api_content_ctags(request: Request, q: str): - """ - Get information about all `Ctags `_-style - symbols defined in a content object. - """ - return api_lookup( - archive.lookup_content_ctags, - q, - notfound_msg="No ctags symbol found for content {}.".format(q), - enrich_fn=utils.enrich_metadata_endpoint, - request=request, - ) - - @api_route( r"/content/(?P[0-9a-z_:]*[0-9a-f]+)/raw/", "api-1-content-raw", checksum_args=["q"], ) @api_doc("/content/raw/") def api_content_raw(request: Request, q: str): """ .. http:get:: /api/1/content/[(hash_type):](hash)/raw/ Get the raw content of a content object (aka a "blob"), as a byte sequence. :param string hash_type: optional parameter specifying which hashing algorithm has been used to compute the content checksum. It can be either ``sha1``, ``sha1_git``, ``sha256`` or ``blake2s256``. If that parameter is not provided, it is assumed that the hashing algorithm used is ``sha1``. :param string hash: hexadecimal representation of the checksum value computed with the specified hashing algorithm. :query string filename: if provided, the downloaded content will get that filename :resheader Content-Type: application/octet-stream :statuscode 200: no error :statuscode 400: an invalid **hash_type** or **hash** has been provided :statuscode 404: requested content can not be found in the archive **Example:** .. parsed-literal:: :swh_web_api:`content/sha1:dc2830a9e72f23c1dfebef4413003221baa5fb62/raw/` """ content_raw = archive.lookup_content_raw(q) if not content_raw: raise NotFoundExc("Content %s is not found." % q) filename = request.query_params.get("filename") if not filename: filename = "content_%s_raw" % q.replace(":", "_") return FileResponse( io.BytesIO(content_raw["data"]), # not copied, as this is never modified filename=filename, content_type="application/octet-stream", as_attachment=True, ) -@api_route(r"/content/symbol/(?P.+)/", "api-1-content-symbol") -@api_doc("/content/symbol/", tags=["hidden"]) -def api_content_symbol(request: Request, q: str): - """Search content objects by `Ctags `_-style - symbol (e.g., function name, data type, method, ...). - - """ - result = {} - last_sha1 = request.query_params.get("last_sha1", None) - per_page = int(request.query_params.get("per_page", "10")) - - def lookup_exp(exp, last_sha1=last_sha1, per_page=per_page): - exp = list(archive.lookup_expression(exp, last_sha1, per_page)) - return exp if exp else None - - symbols = api_lookup( - lookup_exp, - q, - notfound_msg="No indexed raw content match expression '{}'.".format(q), - enrich_fn=functools.partial(utils.enrich_content, top_url=True), - request=request, - ) - - if symbols: - nb_symbols = len(symbols) - - if nb_symbols == per_page: - query_params = {} - new_last_sha1 = symbols[-1]["sha1"] - query_params["last_sha1"] = new_last_sha1 - if request.query_params.get("per_page"): - query_params["per_page"] = per_page - - result["headers"] = { - "link-next": reverse( - "api-1-content-symbol", - url_args={"q": q}, - query_params=query_params, - request=request, - ) - } - - result.update({"results": symbols}) - - return result - - @api_route(r"/content/known/search/", "api-1-content-known", methods=["POST"]) @api_route(r"/content/known/(?P(?!search).+)/", "api-1-content-known") @api_doc("/content/known/", tags=["hidden"]) @format_docstring() def api_check_content_known(request: Request, q: Optional[str] = None): """ .. http:get:: /api/1/content/known/(sha1)[,(sha1), ...,(sha1)]/ Check whether some content(s) (aka "blob(s)") is present in the archive based on its **sha1** checksum. :param string sha1: hexadecimal representation of the **sha1** checksum value for the content to check existence. Multiple values can be provided separated by ','. {common_headers} :>json array search_res: array holding the search result for each provided **sha1** :>json object search_stats: some statistics regarding the number of **sha1** provided and the percentage of those found in the archive :statuscode 200: no error :statuscode 400: an invalid **sha1** has been provided **Example:** .. parsed-literal:: :swh_web_api:`content/known/dc2830a9e72f23c1dfebef4413003221baa5fb62,0c3f19cb47ebfbe643fb19fa94c874d18fa62d12/` """ search_stats = {"nbfiles": 0, "pct": 0} search_res = None queries = [] # GET: Many hash separated values request if q: hashes = q.split(",") for v in hashes: queries.append({"filename": None, "sha1": v}) # POST: Many hash requests in post form submission elif request.method == "POST": data = request.data # Remove potential inputs with no associated value for k, v in data.items(): if v is not None: if k == "q" and len(v) > 0: queries.append({"filename": None, "sha1": v}) elif v != "": queries.append({"filename": k, "sha1": v}) if queries: lookup = archive.lookup_multiple_hashes(queries) result = [] nb_queries = len(queries) for el in lookup: res_d = {"sha1": el["sha1"], "found": el["found"]} if "filename" in el and el["filename"]: res_d["filename"] = el["filename"] result.append(res_d) search_res = result nbfound = len([x for x in lookup if x["found"]]) search_stats["nbfiles"] = nb_queries search_stats["pct"] = int((nbfound / nb_queries) * 100) return {"search_res": search_res, "search_stats": search_stats} @api_route( r"/content/(?P[0-9a-z_:]*[0-9a-f]+)/", "api-1-content", checksum_args=["q"] ) @api_doc("/content/") @format_docstring() def api_content_metadata(request: Request, q: str): """ .. http:get:: /api/1/content/[(hash_type):](hash)/ Get information about a content (aka a "blob") object. In the archive, a content object is identified based on checksum values computed using various hashing algorithms. :param string hash_type: optional parameter specifying which hashing algorithm has been used to compute the content checksum. It can be either ``sha1``, ``sha1_git``, ``sha256`` or ``blake2s256``. If that parameter is not provided, it is assumed that the hashing algorithm used is ``sha1``. :param string hash: hexadecimal representation of the checksum value computed with the specified hashing algorithm. {common_headers} :>json object checksums: object holding the computed checksum values for the requested content :>json string data_url: link to :http:get:`/api/1/content/[(hash_type):](hash)/raw/` for downloading the content raw bytes :>json string filetype_url: link to :http:get:`/api/1/content/[(hash_type):](hash)/filetype/` for getting information about the content MIME type :>json string language_url: link to :http:get:`/api/1/content/[(hash_type):](hash)/language/` for getting information about the programming language used in the content :>json number length: length of the content in bytes :>json string license_url: link to :http:get:`/api/1/content/[(hash_type):](hash)/license/` for getting information about the license of the content :statuscode 200: no error :statuscode 400: an invalid **hash_type** or **hash** has been provided :statuscode 404: requested content can not be found in the archive **Example:** .. parsed-literal:: :swh_web_api:`content/sha1_git:fe95a46679d128ff167b7c55df5d02356c5a1ae1/` """ return api_lookup( archive.lookup_content, q, notfound_msg="Content with {} not found.".format(q), enrich_fn=functools.partial(utils.enrich_content, query_string=q), request=request, ) diff --git a/swh/web/common/archive.py b/swh/web/common/archive.py index 712480ce..41e06a9a 100644 --- a/swh/web/common/archive.py +++ b/swh/web/common/archive.py @@ -1,1477 +1,1429 @@ # Copyright (C) 2015-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from collections import defaultdict import datetime 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.model import 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.storage.interface import OriginVisitWithStatuses 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 = 1000 # 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, lookup_similar_urls: bool = True) -> OriginInfo: """Return information about the origin matching dict origin. Args: origin: origin's dict with 'url' key lookup_similar_urls: if :const:`True`, lookup origin with and without trailing slash in its URL Returns: origin information as dict. """ origin_urls = [origin["url"]] if origin["url"] and lookup_similar_urls: # 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 lookup_origin_snapshots(origin: OriginInfo) -> List[str]: """Return ids of the snapshots of an origin. Args: origin: origin's dict with 'url' key Returns: List of unique snapshot identifiers in hexadecimal format resulting from the visits of the origin. """ return [ snapshot.hex() for snapshot in storage.origin_snapshot_get_all(origin["url"]) ] 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 for field in ("from_directory", "from_revision"): # from_directory when using swh.indexer >= 2.0.0, from_revision otherwise if field in match: match[field] = hashutil.hash_to_hex(match[field]) 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=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=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[OriginVisitWithStatuses]: """Yields the origin origins' visits. Args: origin_url: origin to list visits for last_visit: last visit to lookup from limit: 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_with_statuses( 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): yield converters.from_origin_visit(visit.statuses[-1].to_dict()) def lookup_origin_visit_latest( origin_url: str, require_snapshot: bool = False, type: Optional[str] = None, allowed_statuses: Optional[List[str]] = None, lookup_similar_urls: bool = True, ) -> 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 lookup_similar_urls: if :const:`True`, lookup origin with and without trailing slash in its URL Returns: The origin visit info as dict if found """ # check origin existence in the archive origin_url = lookup_origin( OriginInfo(url=origin_url), lookup_similar_urls=lookup_similar_urls )["url"] 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, lookup_similar_urls: bool = True, ) -> OriginVisitInfo: """Return information about visit visit_id with origin origin. Args: origin: origin concerned by the visit visit_id: the visit identifier to lookup lookup_similar_urls: if :const:`True`, lookup origin with and without trailing slash in its URL Raises: NotFoundExc if no origin visit matching the criteria is found Returns: The dict origin_visit concerned """ # check origin existence in the archive origin_url = lookup_origin( OriginInfo(url=origin_url), lookup_similar_urls=lookup_similar_urls )["url"] 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 origin_visit_find_by_date( origin_url: str, visit_date: datetime.datetime, greater_or_equal: bool = True ) -> Optional[OriginVisitInfo]: """Retrieve origin visit status whose date is most recent than the provided visit_date. Args: origin_url: origin concerned by the visit visit_date: provided visit date greater_or_equal: ensure returned visit has a date greater or equal than the one passed as parameter Returns: The dict origin_visit_status matching the criteria if any. """ visit = storage.origin_visit_find_by_date(origin_url, visit_date) if greater_or_equal and visit and visit.date < visit_date: # when visit is anterior to the provided date, trying to find another one most # recent visits = storage.origin_visit_get( origin_url, page_token=str(visit.visit), limit=1, ).results visit = visits[0] if visits else None if not visit: return None visit_status = storage.origin_visit_status_get_latest(origin_url, visit.visit) 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[ObjectType, 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/tests/api/views/test_content.py b/swh/web/tests/api/views/test_content.py index af29714a..367ab870 100644 --- a/swh/web/tests/api/views/test_content.py +++ b/swh/web/tests/api/views/test_content.py @@ -1,339 +1,251 @@ # Copyright (C) 2015-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import pytest from swh.web.common.utils import reverse -from swh.web.tests.conftest import ctags_json_missing, fossology_missing +from swh.web.tests.conftest import fossology_missing from swh.web.tests.data import random_content from swh.web.tests.utils import ( check_api_get_responses, check_api_post_responses, check_http_get_response, ) def test_api_content_filetype(api_client, indexer_data, content): indexer_data.content_add_mimetype(content["sha1"]) url = reverse( "api-1-content-filetype", url_args={"q": "sha1_git:%s" % content["sha1_git"]} ) rv = check_api_get_responses(api_client, url, status_code=200) content_url = reverse( "api-1-content", url_args={"q": "sha1:%s" % content["sha1"]}, request=rv.wsgi_request, ) expected_data = indexer_data.content_get_mimetype(content["sha1"]) expected_data["content_url"] = content_url assert rv.data == expected_data def test_api_content_filetype_sha_not_found(api_client): unknown_content_ = random_content() url = reverse( "api-1-content-filetype", url_args={"q": "sha1:%s" % unknown_content_["sha1"]} ) rv = check_api_get_responses(api_client, url, status_code=404) assert rv.data == { "exception": "NotFoundExc", "reason": "No filetype information found for content " "sha1:%s." % unknown_content_["sha1"], } def test_api_content_language_sha_not_found(api_client): unknown_content_ = random_content() url = reverse( "api-1-content-language", url_args={"q": "sha1:%s" % unknown_content_["sha1"]} ) rv = check_api_get_responses(api_client, url, status_code=404) assert rv.data == { "exception": "NotFoundExc", "reason": "No language information found for content " "sha1:%s." % unknown_content_["sha1"], } -@pytest.mark.skip # Language indexer is disabled -@pytest.mark.skipif( - ctags_json_missing, reason="requires ctags with json output support" -) -def test_api_content_symbol(api_client, indexer_data, contents_with_ctags): - expected_data = {} - for content_sha1 in contents_with_ctags["sha1s"]: - indexer_data.content_add_ctags(content_sha1) - for ctag in indexer_data.content_get_ctags(content_sha1): - if ctag["name"] == contents_with_ctags["symbol_name"]: - expected_data[content_sha1] = ctag - break - url = reverse( - "api-1-content-symbol", - url_args={"q": contents_with_ctags["symbol_name"]}, - query_params={"per_page": 100}, - ) - rv = check_api_get_responses(api_client, url, status_code=200) - - for entry in rv.data: - content_sha1 = entry["sha1"] - expected_entry = expected_data[content_sha1] - for key, view_name in ( - ("content_url", "api-1-content"), - ("data_url", "api-1-content-raw"), - ("license_url", "api-1-content-license"), - ("language_url", "api-1-content-language"), - ("filetype_url", "api-1-content-filetype"), - ): - expected_entry[key] = reverse( - view_name, - url_args={"q": "sha1:%s" % content_sha1}, - request=rv.wsgi_request, - ) - expected_entry["sha1"] = content_sha1 - del expected_entry["id"] - assert entry == expected_entry - assert "Link" not in rv - - url = reverse( - "api-1-content-symbol", - url_args={"q": contents_with_ctags["symbol_name"]}, - query_params={"per_page": 2}, - ) - - rv = check_api_get_responses(api_client, url, status_code=200) - - next_url = ( - reverse( - "api-1-content-symbol", - url_args={"q": contents_with_ctags["symbol_name"]}, - query_params={"last_sha1": rv.data[1]["sha1"], "per_page": 2}, - request=rv.wsgi_request, - ), - ) - assert rv["Link"] == '<%s>; rel="next"' % next_url - - -def test_api_content_symbol_not_found(api_client): - url = reverse("api-1-content-symbol", url_args={"q": "bar"}) - rv = check_api_get_responses(api_client, url, status_code=404) - assert rv.data == { - "exception": "NotFoundExc", - "reason": "No indexed raw content match expression 'bar'.", - } - assert "Link" not in rv - - -@pytest.mark.skipif( - ctags_json_missing, reason="requires ctags with json output support" -) -def test_api_content_ctags(api_client, indexer_data, content): - indexer_data.content_add_ctags(content["sha1"]) - url = reverse( - "api-1-content-ctags", url_args={"q": "sha1_git:%s" % content["sha1_git"]} - ) - rv = check_api_get_responses(api_client, url, status_code=200) - content_url = reverse( - "api-1-content", - url_args={"q": "sha1:%s" % content["sha1"]}, - request=rv.wsgi_request, - ) - expected_data = list(indexer_data.content_get_ctags(content["sha1"])) - for e in expected_data: - e["content_url"] = content_url - assert rv.data == expected_data - - @pytest.mark.skipif(fossology_missing, reason="requires fossology-nomossa installed") def test_api_content_license(api_client, indexer_data, content): indexer_data.content_add_license(content["sha1"]) url = reverse( "api-1-content-license", url_args={"q": "sha1_git:%s" % content["sha1_git"]} ) rv = check_api_get_responses(api_client, url, status_code=200) content_url = reverse( "api-1-content", url_args={"q": "sha1:%s" % content["sha1"]}, request=rv.wsgi_request, ) expected_data = list(indexer_data.content_get_license(content["sha1"])) for license in expected_data: del license["id"] assert rv.data == { "content_url": content_url, "id": content["sha1"], "facts": expected_data, } def test_api_content_license_sha_not_found(api_client): unknown_content_ = random_content() url = reverse( "api-1-content-license", url_args={"q": "sha1:%s" % unknown_content_["sha1"]} ) rv = check_api_get_responses(api_client, url, status_code=404) assert rv.data == { "exception": "NotFoundExc", "reason": "No license information found for content " "sha1:%s." % unknown_content_["sha1"], } def test_api_content_metadata(api_client, archive_data, content): url = reverse("api-1-content", {"q": "sha1:%s" % content["sha1"]}) rv = check_api_get_responses(api_client, url, status_code=200) expected_data = archive_data.content_get(content["sha1"]) for key, view_name in ( ("data_url", "api-1-content-raw"), ("license_url", "api-1-content-license"), ("language_url", "api-1-content-language"), ("filetype_url", "api-1-content-filetype"), ): expected_data[key] = reverse( view_name, url_args={"q": "sha1:%s" % content["sha1"]}, request=rv.wsgi_request, ) assert rv.data == expected_data def test_api_content_not_found(api_client): unknown_content_ = random_content() url = reverse("api-1-content", url_args={"q": "sha1:%s" % unknown_content_["sha1"]}) rv = check_api_get_responses(api_client, url, status_code=404) assert rv.data == { "exception": "NotFoundExc", "reason": "Content with sha1 checksum equals to %s not found!" % unknown_content_["sha1"], } def test_api_content_raw_ko_not_found(api_client): unknown_content_ = random_content() url = reverse( "api-1-content-raw", url_args={"q": "sha1:%s" % unknown_content_["sha1"]} ) rv = check_api_get_responses(api_client, url, status_code=404) assert rv.data == { "exception": "NotFoundExc", "reason": "Content with sha1 checksum equals to %s not found!" % unknown_content_["sha1"], } def test_api_content_raw_text(api_client, archive_data, content): url = reverse("api-1-content-raw", url_args={"q": "sha1:%s" % content["sha1"]}) rv = check_http_get_response(api_client, url, status_code=200) assert rv["Content-Type"] == "application/octet-stream" assert ( rv["Content-disposition"] == 'attachment; filename="content_sha1_%s_raw"' % content["sha1"] ) expected_data = archive_data.content_get_data(content["sha1"]) assert b"".join(rv.streaming_content) == expected_data["data"] assert int(rv["Content-Length"]) == len(expected_data["data"]) def test_api_content_raw_text_with_filename(api_client, archive_data, content): url = reverse( "api-1-content-raw", url_args={"q": "sha1:%s" % content["sha1"]}, query_params={"filename": "filename.txt"}, ) rv = check_http_get_response(api_client, url, status_code=200) assert rv["Content-disposition"] == 'attachment; filename="filename.txt"' assert rv["Content-Type"] == "application/octet-stream" expected_data = archive_data.content_get_data(content["sha1"]) assert b"".join(rv.streaming_content) == expected_data["data"] assert int(rv["Content-Length"]) == len(expected_data["data"]) @pytest.mark.parametrize( "encoded,expected", [ # From https://datatracker.ietf.org/doc/html/rfc5987#section-3.2.2 ( "%c2%a3%20and%20%e2%82%ac%20rates.txt", "%C2%A3%20and%20%E2%82%AC%20rates.txt", ), ("%A3%20rates.txt", "%EF%BF%BD%20rates.txt"), # found in the wild ( "Th%C3%A9orie%20de%20sant%C3%A9-aide-justice.pdf", "Th%C3%A9orie%20de%20sant%C3%A9-aide-justice.pdf", ), ], ) def test_api_content_raw_text_with_nonascii_filename( api_client, archive_data, content, encoded, expected ): url = reverse( "api-1-content-raw", url_args={"q": "sha1:%s" % content["sha1"]}, ) rv = check_http_get_response( api_client, f"{url}?filename={encoded}", status_code=200 ) # technically, ISO8859-1 is allowed too assert rv["Content-disposition"].isascii(), rv["Content-disposition"] assert rv["Content-disposition"] == ( f"attachment; filename*=utf-8''{expected}" ), rv["Content-disposition"] assert rv["Content-Type"] == "application/octet-stream" expected_data = archive_data.content_get_data(content["sha1"]) assert b"".join(rv.streaming_content) == expected_data["data"] assert int(rv["Content-Length"]) == len(expected_data["data"]) def test_api_check_content_known(api_client, content): url = reverse("api-1-content-known", url_args={"q": content["sha1"]}) rv = check_api_get_responses(api_client, url, status_code=200) assert rv.data == { "search_res": [{"found": True, "sha1": content["sha1"]}], "search_stats": {"nbfiles": 1, "pct": 100.0}, } def test_api_check_content_known_post(api_client, content): url = reverse("api-1-content-known") rv = check_api_post_responses( api_client, url, data={"q": content["sha1"]}, status_code=200 ) assert rv.data == { "search_res": [{"found": True, "sha1": content["sha1"]}], "search_stats": {"nbfiles": 1, "pct": 100.0}, } def test_api_check_content_known_not_found(api_client): unknown_content_ = random_content() url = reverse("api-1-content-known", url_args={"q": unknown_content_["sha1"]}) rv = check_api_get_responses(api_client, url, status_code=200) assert rv.data == { "search_res": [{"found": False, "sha1": unknown_content_["sha1"]}], "search_stats": {"nbfiles": 1, "pct": 0.0}, } def test_api_content_uppercase(api_client, content): url = reverse( "api-1-content-uppercase-checksum", url_args={"q": content["sha1"].upper()} ) rv = check_http_get_response(api_client, url, status_code=302) redirect_url = reverse("api-1-content", url_args={"q": content["sha1"]}) assert rv["location"] == redirect_url diff --git a/swh/web/tests/common/test_archive.py b/swh/web/tests/common/test_archive.py index 2812176c..27cac5c5 100644 --- a/swh/web/tests/common/test_archive.py +++ b/swh/web/tests/common/test_archive.py @@ -1,1253 +1,1194 @@ # Copyright (C) 2015-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from collections import defaultdict import datetime import hashlib import itertools import random from hypothesis import given, settings import pytest from swh.model.from_disk import DentryPerms from swh.model.hashutil import hash_to_bytes, hash_to_hex from swh.model.model import ( Directory, DirectoryEntry, Origin, OriginVisit, OriginVisitStatus, Revision, Snapshot, SnapshotBranch, TargetType, ) from swh.model.swhids import ObjectType from swh.storage.utils import now from swh.web.common import archive from swh.web.common.exc import BadInputExc, NotFoundExc from swh.web.common.typing import OriginInfo, PagedResult -from swh.web.tests.conftest import ctags_json_missing, fossology_missing +from swh.web.tests.conftest import fossology_missing from swh.web.tests.data import random_content, random_sha1 from swh.web.tests.strategies import new_origin, new_revision, visit_dates def test_lookup_multiple_hashes_all_present(contents): input_data = [] expected_output = [] for cnt in contents: input_data.append({"sha1": cnt["sha1"]}) expected_output.append({"sha1": cnt["sha1"], "found": True}) assert archive.lookup_multiple_hashes(input_data) == expected_output def test_lookup_multiple_hashes_some_missing(contents, unknown_contents): input_contents = list(itertools.chain(contents, unknown_contents)) random.shuffle(input_contents) input_data = [] expected_output = [] for cnt in input_contents: input_data.append({"sha1": cnt["sha1"]}) expected_output.append({"sha1": cnt["sha1"], "found": cnt in contents}) assert archive.lookup_multiple_hashes(input_data) == expected_output def test_lookup_hash_does_not_exist(): unknown_content_ = random_content() actual_lookup = archive.lookup_hash("sha1_git:%s" % unknown_content_["sha1_git"]) assert actual_lookup == {"found": None, "algo": "sha1_git"} def test_lookup_hash_exist(archive_data, content): actual_lookup = archive.lookup_hash("sha1:%s" % content["sha1"]) content_metadata = archive_data.content_get(content["sha1"]) assert {"found": content_metadata, "algo": "sha1"} == actual_lookup def test_search_hash_does_not_exist(): unknown_content_ = random_content() actual_lookup = archive.search_hash("sha1_git:%s" % unknown_content_["sha1_git"]) assert {"found": False} == actual_lookup def test_search_hash_exist(content): actual_lookup = archive.search_hash("sha1:%s" % content["sha1"]) assert {"found": True} == actual_lookup -@pytest.mark.skipif( - ctags_json_missing, reason="requires ctags with json output support" -) -def test_lookup_content_ctags(indexer_data, contents_with_ctags): - content_sha1 = random.choice(contents_with_ctags["sha1s"]) - indexer_data.content_add_ctags(content_sha1) - actual_ctags = list(archive.lookup_content_ctags("sha1:%s" % content_sha1)) - - expected_data = list(indexer_data.content_get_ctags(content_sha1)) - for ctag in expected_data: - ctag["id"] = content_sha1 - - assert actual_ctags == expected_data - - -def test_lookup_content_ctags_no_hash(): - unknown_content_ = random_content() - - actual_ctags = list( - archive.lookup_content_ctags("sha1:%s" % unknown_content_["sha1"]) - ) - - assert actual_ctags == [] - - def test_lookup_content_filetype(indexer_data, content): indexer_data.content_add_mimetype(content["sha1"]) actual_filetype = archive.lookup_content_filetype(content["sha1"]) expected_filetype = indexer_data.content_get_mimetype(content["sha1"]) assert actual_filetype == expected_filetype -def test_lookup_expression(indexer_data, contents_with_ctags): - per_page = 10 - expected_ctags = [] - - for content_sha1 in contents_with_ctags["sha1s"]: - if len(expected_ctags) == per_page: - break - indexer_data.content_add_ctags(content_sha1) - for ctag in indexer_data.content_get_ctags(content_sha1): - if len(expected_ctags) == per_page: - break - if ctag["name"] == contents_with_ctags["symbol_name"]: - del ctag["id"] - ctag["sha1"] = content_sha1 - expected_ctags.append(ctag) - - actual_ctags = list( - archive.lookup_expression( - contents_with_ctags["symbol_name"], last_sha1=None, per_page=10 - ) - ) - - assert actual_ctags == expected_ctags - - -def test_lookup_expression_no_result(): - expected_ctags = [] - - actual_ctags = list( - archive.lookup_expression("barfoo", last_sha1=None, per_page=10) - ) - assert actual_ctags == expected_ctags - - @pytest.mark.skipif(fossology_missing, reason="requires fossology-nomossa installed") def test_lookup_content_license(indexer_data, content): indexer_data.content_add_license(content["sha1"]) actual_license = archive.lookup_content_license(content["sha1"]) expected_license = indexer_data.content_get_license(content["sha1"]) assert actual_license == expected_license def test_stat_counters(archive_data): actual_stats = archive.stat_counters() assert actual_stats == archive_data.stat_counters() @given(new_origin(), visit_dates()) def test_lookup_origin_visits(subtest, new_origin, visit_dates): # ensure archive_data fixture will be reset between each hypothesis # example test run @subtest def test_inner(archive_data): archive_data.origin_add([new_origin]) archive_data.origin_visit_add( [ OriginVisit( origin=new_origin.url, date=ts, type="git", ) for ts in visit_dates ] ) actual_origin_visits = list( archive.lookup_origin_visits(new_origin.url, per_page=100) ) expected_visits = archive_data.origin_visit_get(new_origin.url) for expected_visit in expected_visits: expected_visit["origin"] = new_origin.url assert actual_origin_visits == expected_visits @given(new_origin(), visit_dates()) def test_lookup_origin_visit(archive_data, new_origin, visit_dates): archive_data.origin_add([new_origin]) visits = archive_data.origin_visit_add( [ OriginVisit( origin=new_origin.url, date=ts, type="git", ) for ts in visit_dates ] ) visit = random.choice(visits).visit actual_origin_visit = archive.lookup_origin_visit(new_origin.url, visit) expected_visit = dict(archive_data.origin_visit_get_by(new_origin.url, visit)) assert actual_origin_visit == expected_visit @given(new_origin(), visit_dates()) @settings(max_examples=1) def test_origin_visit_find_by_date_no_result(archive_data, new_origin, visit_dates): """No visit registered in storage for an origin should return no visit""" archive_data.origin_add([new_origin]) for visit_date in visit_dates: # No visit yet, so nothing will get returned actual_origin_visit_status = archive.origin_visit_find_by_date( new_origin.url, visit_date ) assert actual_origin_visit_status is None @settings(max_examples=1) @given(new_origin()) def test_origin_visit_find_by_date(archive_data, new_origin): # Add origin and two visits archive_data.origin_add([new_origin]) pivot_date = now() # First visit one hour before pivot date first_visit_date = pivot_date - datetime.timedelta(hours=1) # Second visit two hours after pivot date second_visit_date = pivot_date + datetime.timedelta(hours=2) visits = archive_data.origin_visit_add( [ OriginVisit( origin=new_origin.url, date=visit_date, type="git", ) for visit_date in [first_visit_date, second_visit_date] ] ) # Finalize visits visit_statuses = [] for visit in visits: visit_statuses.append( OriginVisitStatus( origin=new_origin.url, visit=visit.visit, date=visit.date + datetime.timedelta(hours=1), type=visit.type, status="full", snapshot=None, ) ) archive_data.origin_visit_status_add(visit_statuses) # Check correct visit is returned when searching by date for search_date, greater_or_equal, expected_visit in [ (first_visit_date, True, 1), (pivot_date, True, 2), (pivot_date, False, 1), (second_visit_date, True, 2), ]: origin_visit = archive.origin_visit_find_by_date( new_origin.url, search_date, greater_or_equal ) assert origin_visit["visit"] == expected_visit @given(new_origin()) def test_lookup_origin(archive_data, new_origin): archive_data.origin_add([new_origin]) actual_origin = archive.lookup_origin({"url": new_origin.url}) expected_origin = archive_data.origin_get([new_origin.url])[0] assert actual_origin == expected_origin def test_lookup_origin_snapshots(archive_data, origin_with_multiple_visits): origin_url = origin_with_multiple_visits["url"] visits = archive_data.origin_visit_get(origin_url) origin_snapshots = archive.lookup_origin_snapshots(origin_with_multiple_visits) assert set(origin_snapshots) == {v["snapshot"] for v in visits} def test_lookup_release_ko_id_checksum_not_a_sha1(invalid_sha1): with pytest.raises(BadInputExc) as e: archive.lookup_release(invalid_sha1) assert e.match("Invalid checksum") def test_lookup_release_ko_id_checksum_too_long(sha256): with pytest.raises(BadInputExc) as e: archive.lookup_release(sha256) assert e.match("Only sha1_git is supported.") def test_lookup_release_multiple(archive_data, releases): actual_releases = list(archive.lookup_release_multiple(releases)) expected_releases = [] for release_id in releases: release_info = archive_data.release_get(release_id) expected_releases.append(release_info) assert actual_releases == expected_releases def test_lookup_release_multiple_none_found(): unknown_releases_ = [random_sha1(), random_sha1(), random_sha1()] actual_releases = list(archive.lookup_release_multiple(unknown_releases_)) assert actual_releases == [None] * len(unknown_releases_) def test_lookup_directory_with_path_not_found(directory): path = "some/invalid/path/here" with pytest.raises(NotFoundExc) as e: archive.lookup_directory_with_path(directory, path) assert e.match( f"Directory entry with path {path} from root directory {directory} not found" ) def test_lookup_directory_with_path_found(archive_data, directory): directory_content = archive_data.directory_ls(directory) directory_entry = random.choice(directory_content) path = directory_entry["name"] actual_result = archive.lookup_directory_with_path(directory, path) assert actual_result == directory_entry def test_lookup_release(archive_data, release): actual_release = archive.lookup_release(release) assert actual_release == archive_data.release_get(release) def test_lookup_revision_with_context_ko_not_a_sha1(revision, invalid_sha1, sha256): sha1_git_root = revision sha1_git = invalid_sha1 with pytest.raises(BadInputExc) as e: archive.lookup_revision_with_context(sha1_git_root, sha1_git) assert e.match("Invalid checksum query string") sha1_git = sha256 with pytest.raises(BadInputExc) as e: archive.lookup_revision_with_context(sha1_git_root, sha1_git) assert e.match("Only sha1_git is supported") def test_lookup_revision_with_context_ko_sha1_git_does_not_exist( revision, unknown_revision ): sha1_git_root = revision sha1_git = unknown_revision with pytest.raises(NotFoundExc) as e: archive.lookup_revision_with_context(sha1_git_root, sha1_git) assert e.match("Revision %s not found" % sha1_git) def test_lookup_revision_with_context_ko_root_sha1_git_does_not_exist( revision, unknown_revision ): sha1_git_root = unknown_revision sha1_git = revision with pytest.raises(NotFoundExc) as e: archive.lookup_revision_with_context(sha1_git_root, sha1_git) assert e.match("Revision root %s not found" % sha1_git_root) def test_lookup_revision_with_context(archive_data, ancestor_revisions): sha1_git = ancestor_revisions["sha1_git"] root_sha1_git = ancestor_revisions["sha1_git_root"] for sha1_git_root in (root_sha1_git, {"id": hash_to_bytes(root_sha1_git)}): actual_revision = archive.lookup_revision_with_context(sha1_git_root, sha1_git) children = [] for rev in archive_data.revision_log(root_sha1_git): for p_rev in rev["parents"]: p_rev_hex = hash_to_hex(p_rev) if p_rev_hex == sha1_git: children.append(rev["id"]) expected_revision = archive_data.revision_get(sha1_git) expected_revision["children"] = children assert actual_revision == expected_revision def test_lookup_revision_with_context_ko(non_ancestor_revisions): sha1_git = non_ancestor_revisions["sha1_git"] root_sha1_git = non_ancestor_revisions["sha1_git_root"] with pytest.raises(NotFoundExc) as e: archive.lookup_revision_with_context(root_sha1_git, sha1_git) assert e.match("Revision %s is not an ancestor of %s" % (sha1_git, root_sha1_git)) def test_lookup_directory_with_revision_not_found(): unknown_revision_ = random_sha1() with pytest.raises(NotFoundExc) as e: archive.lookup_directory_with_revision(unknown_revision_) assert e.match("Revision %s not found" % unknown_revision_) @given(new_revision()) def test_lookup_directory_with_revision_unknown_content(archive_data, new_revision): unknown_content_ = random_content() dir_path = "README.md" # A directory that points to unknown content dir = Directory( entries=( DirectoryEntry( name=bytes(dir_path.encode("utf-8")), type="file", target=hash_to_bytes(unknown_content_["sha1_git"]), perms=DentryPerms.content, ), ) ) # Create a revision that points to a directory # Which points to unknown content new_revision = new_revision.to_dict() new_revision["directory"] = dir.id del new_revision["id"] new_revision = Revision.from_dict(new_revision) # Add the directory and revision in mem archive_data.directory_add([dir]) archive_data.revision_add([new_revision]) new_revision_id = hash_to_hex(new_revision.id) with pytest.raises(NotFoundExc) as e: archive.lookup_directory_with_revision(new_revision_id, dir_path) assert e.match("Content not found for revision %s" % new_revision_id) def test_lookup_directory_with_revision_ko_path_to_nowhere(revision): invalid_path = "path/to/something/unknown" with pytest.raises(NotFoundExc) as e: archive.lookup_directory_with_revision(revision, invalid_path) assert e.match("Directory or File") assert e.match(invalid_path) assert e.match("revision %s" % revision) assert e.match("not found") def test_lookup_directory_with_revision_submodules( archive_data, revision_with_submodules ): rev_sha1_git = revision_with_submodules["rev_sha1_git"] rev_dir_path = revision_with_submodules["rev_dir_rev_path"] actual_data = archive.lookup_directory_with_revision(rev_sha1_git, rev_dir_path) revision = archive_data.revision_get(revision_with_submodules["rev_sha1_git"]) directory = archive_data.directory_ls(revision["directory"]) rev_entry = next(e for e in directory if e["name"] == rev_dir_path) expected_data = { "content": archive_data.revision_get(rev_entry["target"]), "path": rev_dir_path, "revision": rev_sha1_git, "type": "rev", } assert actual_data == expected_data def test_lookup_directory_with_revision_without_path(archive_data, revision): actual_directory_entries = archive.lookup_directory_with_revision(revision) revision_data = archive_data.revision_get(revision) expected_directory_entries = archive_data.directory_ls(revision_data["directory"]) assert actual_directory_entries["type"] == "dir" assert actual_directory_entries["content"] == expected_directory_entries def test_lookup_directory_with_revision_with_path(archive_data, revision): rev_data = archive_data.revision_get(revision) dir_entries = [ e for e in archive_data.directory_ls(rev_data["directory"]) if e["type"] in ("file", "dir") ] expected_dir_entry = random.choice(dir_entries) actual_dir_entry = archive.lookup_directory_with_revision( revision, expected_dir_entry["name"] ) assert actual_dir_entry["type"] == expected_dir_entry["type"] assert actual_dir_entry["revision"] == revision assert actual_dir_entry["path"] == expected_dir_entry["name"] if actual_dir_entry["type"] == "file": del actual_dir_entry["content"]["checksums"]["blake2s256"] for key in ("checksums", "status", "length"): assert actual_dir_entry["content"][key] == expected_dir_entry[key] else: sub_dir_entries = archive_data.directory_ls(expected_dir_entry["target"]) assert actual_dir_entry["content"] == sub_dir_entries def test_lookup_directory_with_revision_with_path_to_file_and_data( archive_data, revision ): rev_data = archive_data.revision_get(revision) dir_entries = [ e for e in archive_data.directory_ls(rev_data["directory"]) if e["type"] == "file" ] expected_dir_entry = random.choice(dir_entries) expected_data = archive_data.content_get_data( expected_dir_entry["checksums"]["sha1"] ) actual_dir_entry = archive.lookup_directory_with_revision( revision, expected_dir_entry["name"], with_data=True ) assert actual_dir_entry["type"] == expected_dir_entry["type"] assert actual_dir_entry["revision"] == revision assert actual_dir_entry["path"] == expected_dir_entry["name"] del actual_dir_entry["content"]["checksums"]["blake2s256"] for key in ("checksums", "status", "length"): assert actual_dir_entry["content"][key] == expected_dir_entry[key] assert actual_dir_entry["content"]["data"] == expected_data["data"] def test_lookup_revision(archive_data, revision): actual_revision = archive.lookup_revision(revision) assert actual_revision == archive_data.revision_get(revision) @given(new_revision()) def test_lookup_revision_invalid_msg(archive_data, new_revision): new_revision = new_revision.to_dict() new_revision["message"] = b"elegant fix for bug \xff" archive_data.revision_add([Revision.from_dict(new_revision)]) revision = archive.lookup_revision(hash_to_hex(new_revision["id"])) assert revision["message"] == "elegant fix for bug \\xff" assert "message" in revision["decoding_failures"] @given(new_revision()) def test_lookup_revision_msg_ok(archive_data, new_revision): archive_data.revision_add([new_revision]) revision_message = archive.lookup_revision_message(hash_to_hex(new_revision.id)) assert revision_message == {"message": new_revision.message} def test_lookup_revision_msg_no_rev(): unknown_revision_ = random_sha1() with pytest.raises(NotFoundExc) as e: archive.lookup_revision_message(unknown_revision_) assert e.match("Revision with sha1_git %s not found." % unknown_revision_) def test_lookup_revision_multiple(archive_data, revisions): actual_revisions = list(archive.lookup_revision_multiple(revisions)) expected_revisions = [] for rev in revisions: expected_revisions.append(archive_data.revision_get(rev)) assert actual_revisions == expected_revisions def test_lookup_revision_multiple_none_found(): unknown_revisions_ = [random_sha1(), random_sha1(), random_sha1()] actual_revisions = list(archive.lookup_revision_multiple(unknown_revisions_)) assert actual_revisions == [None] * len(unknown_revisions_) def test_lookup_revision_log(archive_data, revision): actual_revision_log = list(archive.lookup_revision_log(revision, limit=25)) expected_revision_log = archive_data.revision_log(revision, limit=25) assert actual_revision_log == expected_revision_log def _get_origin_branches(archive_data, origin): origin_visit = archive_data.origin_visit_get(origin["url"])[-1] snapshot = archive_data.snapshot_get(origin_visit["snapshot"]) branches = { k: v for (k, v) in snapshot["branches"].items() if v["target_type"] == "revision" } return branches def test_lookup_revision_log_by(archive_data, origin): branches = _get_origin_branches(archive_data, origin) branch_name = random.choice(list(branches.keys())) actual_log = list( archive.lookup_revision_log_by(origin["url"], branch_name, None, limit=25) ) expected_log = archive_data.revision_log(branches[branch_name]["target"], limit=25) assert actual_log == expected_log def test_lookup_revision_log_by_notfound(origin): with pytest.raises(NotFoundExc): archive.lookup_revision_log_by( origin["url"], "unknown_branch_name", None, limit=100 ) def test_lookup_content_raw_not_found(): unknown_content_ = random_content() with pytest.raises(NotFoundExc) as e: archive.lookup_content_raw("sha1:" + unknown_content_["sha1"]) assert e.match( "Content with %s checksum equals to %s not found!" % ("sha1", unknown_content_["sha1"]) ) def test_lookup_content_raw(archive_data, content): actual_content = archive.lookup_content_raw("sha256:%s" % content["sha256"]) expected_content = archive_data.content_get_data(content["sha1"]) assert actual_content == expected_content def test_lookup_empty_content_raw(empty_content): content_raw = archive.lookup_content_raw(f"sha1_git:{empty_content['sha1_git']}") assert content_raw["data"] == b"" def test_lookup_content_not_found(): unknown_content_ = random_content() with pytest.raises(NotFoundExc) as e: archive.lookup_content("sha1:%s" % unknown_content_["sha1"]) assert e.match( "Content with %s checksum equals to %s not found!" % ("sha1", unknown_content_["sha1"]) ) def test_lookup_content_with_sha1(archive_data, content): actual_content = archive.lookup_content(f"sha1:{content['sha1']}") expected_content = archive_data.content_get(content["sha1"]) assert actual_content == expected_content def test_lookup_content_with_sha256(archive_data, content): actual_content = archive.lookup_content(f"sha256:{content['sha256']}") expected_content = archive_data.content_get(content["sha1"]) assert actual_content == expected_content def test_lookup_directory_bad_checksum(): with pytest.raises(BadInputExc): archive.lookup_directory("directory_id") def test_lookup_directory_not_found(): unknown_directory_ = random_sha1() with pytest.raises(NotFoundExc) as e: archive.lookup_directory(unknown_directory_) assert e.match("Directory with sha1_git %s not found" % unknown_directory_) def test_lookup_directory(archive_data, directory): actual_directory_ls = list(archive.lookup_directory(directory)) expected_directory_ls = archive_data.directory_ls(directory) assert actual_directory_ls == expected_directory_ls def test_lookup_directory_empty(empty_directory): actual_directory_ls = list(archive.lookup_directory(empty_directory)) assert actual_directory_ls == [] def test_lookup_revision_by_nothing_found(origin): with pytest.raises(NotFoundExc): archive.lookup_revision_by(origin["url"], "invalid-branch-name") def test_lookup_revision_by(archive_data, origin): branches = _get_origin_branches(archive_data, origin) branch_name = random.choice(list(branches.keys())) actual_revision = archive.lookup_revision_by(origin["url"], branch_name) expected_revision = archive_data.revision_get(branches[branch_name]["target"]) assert actual_revision == expected_revision def test_lookup_revision_with_context_by_ko(origin, revision): with pytest.raises(NotFoundExc): archive.lookup_revision_with_context_by( origin["url"], "invalid-branch-name", None, revision ) def test_lookup_revision_with_context_by(archive_data, origin): branches = _get_origin_branches(archive_data, origin) branch_name = random.choice(list(branches.keys())) root_rev = branches[branch_name]["target"] root_rev_log = archive_data.revision_log(root_rev) children = defaultdict(list) for rev in root_rev_log: for rev_p in rev["parents"]: children[rev_p].append(rev["id"]) rev = root_rev_log[-1]["id"] actual_root_rev, actual_rev = archive.lookup_revision_with_context_by( origin["url"], branch_name, None, rev ) expected_root_rev = archive_data.revision_get(root_rev) expected_rev = archive_data.revision_get(rev) expected_rev["children"] = children[rev] assert actual_root_rev == expected_root_rev assert actual_rev == expected_rev def test_lookup_revision_through_ko_not_implemented(): with pytest.raises(NotImplementedError): archive.lookup_revision_through({"something-unknown": 10}) def test_lookup_revision_through_with_context_by(archive_data, origin): branches = _get_origin_branches(archive_data, origin) branch_name = random.choice(list(branches.keys())) root_rev = branches[branch_name]["target"] root_rev_log = archive_data.revision_log(root_rev) rev = root_rev_log[-1]["id"] assert archive.lookup_revision_through( { "origin_url": origin["url"], "branch_name": branch_name, "ts": None, "sha1_git": rev, } ) == archive.lookup_revision_with_context_by(origin["url"], branch_name, None, rev) def test_lookup_revision_through_with_revision_by(archive_data, origin): branches = _get_origin_branches(archive_data, origin) branch_name = random.choice(list(branches.keys())) assert archive.lookup_revision_through( { "origin_url": origin["url"], "branch_name": branch_name, "ts": None, } ) == archive.lookup_revision_by(origin["url"], branch_name, None) def test_lookup_revision_through_with_context(ancestor_revisions): sha1_git = ancestor_revisions["sha1_git"] sha1_git_root = ancestor_revisions["sha1_git_root"] assert archive.lookup_revision_through( { "sha1_git_root": sha1_git_root, "sha1_git": sha1_git, } ) == archive.lookup_revision_with_context(sha1_git_root, sha1_git) def test_lookup_revision_through_with_revision(revision): assert archive.lookup_revision_through( {"sha1_git": revision} ) == archive.lookup_revision(revision) def test_lookup_directory_through_revision_ko_not_found(revision): with pytest.raises(NotFoundExc): archive.lookup_directory_through_revision( {"sha1_git": revision}, "some/invalid/path" ) def test_lookup_directory_through_revision_ok(archive_data, revision): rev_data = archive_data.revision_get(revision) dir_entries = [ e for e in archive_data.directory_ls(rev_data["directory"]) if e["type"] == "file" ] dir_entry = random.choice(dir_entries) assert archive.lookup_directory_through_revision( {"sha1_git": revision}, dir_entry["name"] ) == (revision, archive.lookup_directory_with_revision(revision, dir_entry["name"])) def test_lookup_directory_through_revision_ok_with_data(archive_data, revision): rev_data = archive_data.revision_get(revision) dir_entries = [ e for e in archive_data.directory_ls(rev_data["directory"]) if e["type"] == "file" ] dir_entry = random.choice(dir_entries) assert archive.lookup_directory_through_revision( {"sha1_git": revision}, dir_entry["name"], with_data=True ) == ( revision, archive.lookup_directory_with_revision( revision, dir_entry["name"], with_data=True ), ) def test_lookup_known_objects( archive_data, content, directory, release, revision, snapshot ): expected = archive_data.content_find(content) assert archive.lookup_object(ObjectType.CONTENT, content["sha1_git"]) == expected expected = archive_data.directory_get(directory) assert archive.lookup_object(ObjectType.DIRECTORY, directory) == expected expected = archive_data.release_get(release) assert archive.lookup_object(ObjectType.RELEASE, release) == expected expected = archive_data.revision_get(revision) assert archive.lookup_object(ObjectType.REVISION, revision) == expected expected = {**archive_data.snapshot_get(snapshot), "next_branch": None} assert archive.lookup_object(ObjectType.SNAPSHOT, snapshot) == expected def test_lookup_unknown_objects( unknown_content, unknown_directory, unknown_release, unknown_revision, unknown_snapshot, ): with pytest.raises(NotFoundExc) as e: archive.lookup_object(ObjectType.CONTENT, unknown_content["sha1_git"]) assert e.match(r"Content.*not found") with pytest.raises(NotFoundExc) as e: archive.lookup_object(ObjectType.DIRECTORY, unknown_directory) assert e.match(r"Directory.*not found") with pytest.raises(NotFoundExc) as e: archive.lookup_object(ObjectType.RELEASE, unknown_release) assert e.match(r"Release.*not found") with pytest.raises(NotFoundExc) as e: archive.lookup_object(ObjectType.REVISION, unknown_revision) assert e.match(r"Revision.*not found") with pytest.raises(NotFoundExc) as e: archive.lookup_object(ObjectType.SNAPSHOT, unknown_snapshot) assert e.match(r"Snapshot.*not found") def test_lookup_invalid_objects(invalid_sha1): with pytest.raises(BadInputExc) as e: archive.lookup_object(ObjectType.CONTENT, invalid_sha1) assert e.match("Invalid hash") with pytest.raises(BadInputExc) as e: archive.lookup_object(ObjectType.DIRECTORY, invalid_sha1) assert e.match("Invalid checksum") with pytest.raises(BadInputExc) as e: archive.lookup_object(ObjectType.RELEASE, invalid_sha1) assert e.match("Invalid checksum") with pytest.raises(BadInputExc) as e: archive.lookup_object(ObjectType.REVISION, invalid_sha1) assert e.match("Invalid checksum") with pytest.raises(BadInputExc) as e: archive.lookup_object(ObjectType.SNAPSHOT, invalid_sha1) assert e.match("Invalid checksum") def test_lookup_missing_hashes_non_present(): missing_cnt = random_sha1() missing_dir = random_sha1() missing_rev = random_sha1() missing_rel = random_sha1() missing_snp = random_sha1() grouped_swhids = { ObjectType.CONTENT: [hash_to_bytes(missing_cnt)], ObjectType.DIRECTORY: [hash_to_bytes(missing_dir)], ObjectType.REVISION: [hash_to_bytes(missing_rev)], ObjectType.RELEASE: [hash_to_bytes(missing_rel)], ObjectType.SNAPSHOT: [hash_to_bytes(missing_snp)], } actual_result = archive.lookup_missing_hashes(grouped_swhids) assert actual_result == { missing_cnt, missing_dir, missing_rev, missing_rel, missing_snp, } def test_lookup_missing_hashes_some_present(content, directory): missing_rev = random_sha1() missing_rel = random_sha1() missing_snp = random_sha1() grouped_swhids = { ObjectType.CONTENT: [hash_to_bytes(content["sha1_git"])], ObjectType.DIRECTORY: [hash_to_bytes(directory)], ObjectType.REVISION: [hash_to_bytes(missing_rev)], ObjectType.RELEASE: [hash_to_bytes(missing_rel)], ObjectType.SNAPSHOT: [hash_to_bytes(missing_snp)], } actual_result = archive.lookup_missing_hashes(grouped_swhids) assert actual_result == {missing_rev, missing_rel, missing_snp} def test_lookup_origin_extra_trailing_slash(origin): origin_info = archive.lookup_origin({"url": f"{origin['url']}/"}) assert origin_info["url"] == origin["url"] def test_lookup_origin_missing_trailing_slash(archive_data): deb_origin = Origin(url="http://snapshot.debian.org/package/r-base/") archive_data.origin_add([deb_origin]) origin_info = archive.lookup_origin({"url": deb_origin.url[:-1]}) assert origin_info["url"] == deb_origin.url def test_lookup_origin_single_slash_after_protocol(archive_data): origin_url = "http://snapshot.debian.org/package/r-base/" malformed_origin_url = "http:/snapshot.debian.org/package/r-base/" archive_data.origin_add([Origin(url=origin_url)]) origin_info = archive.lookup_origin({"url": malformed_origin_url}) assert origin_info["url"] == origin_url @given(new_origin()) def test_lookup_origins_get_by_sha1s(origin, unknown_origin): hasher = hashlib.sha1() hasher.update(origin["url"].encode("utf-8")) origin_info = OriginInfo(url=origin["url"]) origin_sha1 = hasher.hexdigest() hasher = hashlib.sha1() hasher.update(unknown_origin.url.encode("utf-8")) unknown_origin_sha1 = hasher.hexdigest() origins = list(archive.lookup_origins_by_sha1s([origin_sha1])) assert origins == [origin_info] origins = list(archive.lookup_origins_by_sha1s([origin_sha1, origin_sha1])) assert origins == [origin_info, origin_info] origins = list(archive.lookup_origins_by_sha1s([origin_sha1, unknown_origin_sha1])) assert origins == [origin_info, None] def test_search_origin(origin): results = archive.search_origin(url_pattern=origin["url"])[0] assert results == [{"url": origin["url"]}] def test_search_origin_use_ql(mocker, origin): ORIGIN = [{"url": origin["url"]}] mock_archive_search = mocker.patch("swh.web.common.archive.search") mock_archive_search.origin_search.return_value = PagedResult( results=ORIGIN, next_page_token=None, ) query = f"origin = '{origin['url']}'" results = archive.search_origin(url_pattern=query, use_ql=True)[0] assert results == ORIGIN mock_archive_search.origin_search.assert_called_with( query=query, page_token=None, with_visit=False, visit_types=None, limit=50 ) def test_lookup_snapshot_sizes(archive_data, snapshot): branches = archive_data.snapshot_get(snapshot)["branches"] expected_sizes = { "alias": 0, "release": 0, "revision": 0, } for branch_name, branch_info in branches.items(): if branch_info is not None: expected_sizes[branch_info["target_type"]] += 1 assert archive.lookup_snapshot_sizes(snapshot) == expected_sizes def test_lookup_snapshot_sizes_with_filtering(archive_data, revision): rev_id = hash_to_bytes(revision) snapshot = Snapshot( branches={ b"refs/heads/master": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), b"refs/heads/incoming": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), b"refs/pull/1": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), b"refs/pull/2": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), }, ) archive_data.snapshot_add([snapshot]) expected_sizes = {"alias": 0, "release": 0, "revision": 2} assert ( archive.lookup_snapshot_sizes( snapshot.id.hex(), branch_name_exclude_prefix="refs/pull/" ) == expected_sizes ) def test_lookup_snapshot_alias(snapshot): resolved_alias = archive.lookup_snapshot_alias(snapshot, "HEAD") assert resolved_alias is not None assert resolved_alias["target_type"] == "revision" assert resolved_alias["target"] is not None def test_lookup_snapshot_missing(revision): with pytest.raises(NotFoundExc): archive.lookup_snapshot(revision) def test_lookup_snapshot_empty_branch_list(archive_data, revision): rev_id = hash_to_bytes(revision) snapshot = Snapshot( branches={ b"refs/heads/master": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), }, ) archive_data.snapshot_add([snapshot]) # FIXME; This test will change once the inconsistency in storage is fixed # postgres backend returns None in case of a missing branch whereas the # in-memory implementation (used in tests) returns a data structure; # hence the inconsistency branches = archive.lookup_snapshot( hash_to_hex(snapshot.id), branch_name_include_substring="non-existing", )["branches"] assert not branches def test_lookup_snapshot_branch_names_filtering(archive_data, revision): rev_id = hash_to_bytes(revision) snapshot = Snapshot( branches={ b"refs/heads/master": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), b"refs/heads/incoming": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), b"refs/pull/1": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), b"refs/pull/2": SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), "non_ascii_name_é".encode(): SnapshotBranch( target=rev_id, target_type=TargetType.REVISION, ), }, ) archive_data.snapshot_add([snapshot]) for include_pattern, exclude_prefix, nb_results in ( ("pull", None, 2), ("incoming", None, 1), ("é", None, 1), (None, "refs/heads/", 3), ("refs", "refs/heads/master", 3), ): branches = archive.lookup_snapshot( hash_to_hex(snapshot.id), branch_name_include_substring=include_pattern, branch_name_exclude_prefix=exclude_prefix, )["branches"] assert len(branches) == nb_results for branch_name in branches: if include_pattern: assert include_pattern in branch_name if exclude_prefix: assert not branch_name.startswith(exclude_prefix) def test_lookup_snapshot_branch_names_filtering_paginated( archive_data, directory, revision ): pattern = "foo" nb_branches_by_target_type = 10 branches = {} for i in range(nb_branches_by_target_type): branches[f"branch/directory/bar{i}".encode()] = SnapshotBranch( target=hash_to_bytes(directory), target_type=TargetType.DIRECTORY, ) branches[f"branch/revision/bar{i}".encode()] = SnapshotBranch( target=hash_to_bytes(revision), target_type=TargetType.REVISION, ) branches[f"branch/directory/{pattern}{i}".encode()] = SnapshotBranch( target=hash_to_bytes(directory), target_type=TargetType.DIRECTORY, ) branches[f"branch/revision/{pattern}{i}".encode()] = SnapshotBranch( target=hash_to_bytes(revision), target_type=TargetType.REVISION, ) snapshot = Snapshot(branches=branches) archive_data.snapshot_add([snapshot]) branches_count = nb_branches_by_target_type // 2 for target_type in ( ObjectType.DIRECTORY.name.lower(), ObjectType.REVISION.name.lower(), ): partial_branches = archive.lookup_snapshot( hash_to_hex(snapshot.id), target_types=[target_type], branches_count=branches_count, branch_name_include_substring=pattern, ) branches = partial_branches["branches"] assert len(branches) == branches_count for branch_name, branch_data in branches.items(): assert pattern in branch_name assert branch_data["target_type"] == target_type for i in range(branches_count): assert f"branch/{target_type}/{pattern}{i}" in branches assert ( partial_branches["next_branch"] == f"branch/{target_type}/{pattern}{branches_count}" ) partial_branches = archive.lookup_snapshot( hash_to_hex(snapshot.id), target_types=[target_type], branches_from=partial_branches["next_branch"], branch_name_include_substring=pattern, ) branches = partial_branches["branches"] assert len(branches) == branches_count for branch_name, branch_data in branches.items(): assert pattern in branch_name assert branch_data["target_type"] == target_type for i in range(branches_count, 2 * branches_count): assert f"branch/{target_type}/{pattern}{i}" in branches assert partial_branches["next_branch"] is None diff --git a/swh/web/tests/common/test_utils.py b/swh/web/tests/common/test_utils.py index e6a980d1..14ee5008 100644 --- a/swh/web/tests/common/test_utils.py +++ b/swh/web/tests/common/test_utils.py @@ -1,393 +1,392 @@ # Copyright (C) 2017-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from base64 import b64encode import datetime import math import sys from urllib.parse import quote import pytest from django.test.utils import override_settings from django.urls import re_path as url from django.urls.exceptions import NoReverseMatch from swh.web.common import utils from swh.web.common.exc import BadInputExc from swh.web.config import SWH_WEB_SERVER_NAME, SWH_WEB_STAGING_SERVER_NAMES, get_config def test_shorten_path_noop(): noops = ["/api/", "/browse/", "/content/symbol/foobar/"] for noop in noops: assert utils.shorten_path(noop) == noop def test_shorten_path_sha1(): sha1 = "aafb16d69fd30ff58afdd69036a26047f3aebdc6" short_sha1 = sha1[:8] + "..." templates = [ "/api/1/content/sha1:%s/", "/api/1/content/sha1_git:%s/", "/api/1/directory/%s/", - "/api/1/content/sha1:%s/ctags/", ] for template in templates: assert utils.shorten_path(template % sha1) == template % short_sha1 def test_shorten_path_sha256(): sha256 = "aafb16d69fd30ff58afdd69036a26047" "213add102934013a014dfca031c41aef" short_sha256 = sha256[:8] + "..." templates = [ "/api/1/content/sha256:%s/", "/api/1/directory/%s/", "/api/1/content/sha256:%s/filetype/", ] for template in templates: assert utils.shorten_path(template % sha256) == template % short_sha256 @pytest.mark.parametrize( "input_timestamp, output_date", [ ( "2016-01-12", datetime.datetime(2016, 1, 12, 0, 0, tzinfo=datetime.timezone.utc), ), ( "2016-01-12T09:19:12+0100", datetime.datetime(2016, 1, 12, 8, 19, 12, tzinfo=datetime.timezone.utc), ), ( "2007-01-14T20:34:22Z", datetime.datetime(2007, 1, 14, 20, 34, 22, tzinfo=datetime.timezone.utc), ), ], ) def test_parse_iso8601_date_to_utc_ok(input_timestamp, output_date): assert utils.parse_iso8601_date_to_utc(input_timestamp) == output_date @pytest.mark.parametrize( "invalid_iso8601_timestamp", ["Today is January 1, 2047 at 8:21:00AM", "1452591542"] ) def test_parse_iso8601_date_to_utc_ko(invalid_iso8601_timestamp): with pytest.raises(BadInputExc): utils.parse_iso8601_date_to_utc(invalid_iso8601_timestamp) def test_format_utc_iso_date(): assert ( utils.format_utc_iso_date("2017-05-04T13:27:13+02:00") == "04 May 2017, 11:27:13 UTC" ) def test_gen_path_info(): input_path = "/home/user/swh-environment/swh-web/" expected_result = [ {"name": "home", "path": "home"}, {"name": "user", "path": "home/user"}, {"name": "swh-environment", "path": "home/user/swh-environment"}, {"name": "swh-web", "path": "home/user/swh-environment/swh-web"}, ] path_info = utils.gen_path_info(input_path) assert path_info == expected_result input_path = "home/user/swh-environment/swh-web" path_info = utils.gen_path_info(input_path) assert path_info == expected_result def test_rst_to_html(): rst = ( "Section\n" "=======\n\n" "**Some strong text**\n\n" "* This is a bulleted list.\n" "* It has two items, the second\n" " item uses two lines.\n" "\n" "1. This is a numbered list.\n" "2. It has two items too.\n" "\n" "#. This is a numbered list.\n" "#. It has two items too.\n" ) expected_html = ( '

Section

\n' "

Some strong text

\n" '
    \n' "

  • This is a bulleted list.

    • \n" "

  • It has two items, the second\n" "item uses two lines.

    • \n" "
\n" '
    \n' "

  1. This is a numbered list.

    2. \n" "

  2. It has two items too.

    3. \n" "

  3. This is a numbered list.

    4. \n" "

  4. It has two items too.

    5. \n" "
\n" "
" ) assert utils.rst_to_html(rst) == expected_html def sample_test_view(request, string, number): pass def sample_test_view_no_url_args(request): pass urlpatterns = [ url( r"^sample/test/(?P.+)/view/(?P[0-9]+)/$", sample_test_view, name="sample-test-view", ), url( r"^sample/test/view/no/url/args/$", sample_test_view_no_url_args, name="sample-test-view-no-url-args", ), ] @override_settings(ROOT_URLCONF=__name__) def test_reverse_url_args_only_ok(): string = "foo" number = 55 url = utils.reverse( "sample-test-view", url_args={"string": string, "number": number} ) assert url == f"/sample/test/{string}/view/{number}/" @override_settings(ROOT_URLCONF=__name__) def test_reverse_url_args_only_ko(): string = "foo" with pytest.raises(NoReverseMatch): utils.reverse("sample-test-view", url_args={"string": string, "number": string}) @override_settings(ROOT_URLCONF=__name__) def test_reverse_no_url_args(): url = utils.reverse("sample-test-view-no-url-args") assert url == "/sample/test/view/no/url/args/" @override_settings(ROOT_URLCONF=__name__) def test_reverse_query_params_only(): start = 0 scope = "foo" url = utils.reverse( "sample-test-view-no-url-args", query_params={"start": start, "scope": scope} ) assert url == f"/sample/test/view/no/url/args/?scope={scope}&start={start}" url = utils.reverse( "sample-test-view-no-url-args", query_params={"start": start, "scope": None} ) assert url == f"/sample/test/view/no/url/args/?start={start}" @override_settings(ROOT_URLCONF=__name__) def test_reverse_query_params_encode(): libname = "libstc++" url = utils.reverse( "sample-test-view-no-url-args", query_params={"libname": libname} ) assert url == f"/sample/test/view/no/url/args/?libname={quote(libname, safe='/;:')}" @override_settings(ROOT_URLCONF=__name__) def test_reverse_url_args_query_params(): string = "foo" number = 55 start = 10 scope = "bar" url = utils.reverse( "sample-test-view", url_args={"string": string, "number": number}, query_params={"start": start, "scope": scope}, ) assert url == f"/sample/test/{string}/view/{number}/?scope={scope}&start={start}" @override_settings(ROOT_URLCONF=__name__) def test_reverse_absolute_uri(request_factory): request = request_factory.get(utils.reverse("sample-test-view-no-url-args")) url = utils.reverse("sample-test-view-no-url-args", request=request) assert url == f"http://{request.META['SERVER_NAME']}/sample/test/view/no/url/args/" def test_get_deposits_list(requests_mock): deposits_data = { "count": 2, "results": [ { "check_task_id": "351820217", "client": 2, "collection": 1, "complete_date": "2021-01-21T07:52:19.919312Z", "external_id": "hal-03116143", "id": 1412, "load_task_id": "351820260", "origin_url": "https://hal.archives-ouvertes.fr/hal-03116143", "parent": None, "reception_date": "2021-01-21T07:52:19.471019Z", "status": "done", "status_detail": None, "swhid": "swh:1:dir:f25157ad1b13cb20ac3457d4f6756b49ac63d079", }, { "check_task_id": "381576507", "client": 2, "collection": 1, "complete_date": "2021-07-07T08:00:44.726676Z", "external_id": "hal-03275052", "id": 1693, "load_task_id": "381576508", "origin_url": "https://hal.archives-ouvertes.fr/hal-03275052", "parent": None, "reception_date": "2021-07-07T08:00:44.327661Z", "status": "done", "status_detail": None, "swhid": "swh:1:dir:825fa96d1810177ec08a772ffa5bd34bbd08b89c", }, ], } config = get_config()["deposit"] private_api_url = config["private_api_url"].rstrip("/") + "/" deposits_list_url = private_api_url + "deposits" basic_auth_payload = ( config["private_api_user"] + ":" + config["private_api_password"] ).encode() requests_mock.get( deposits_list_url, json=deposits_data, request_headers={ "Authorization": f"Basic {b64encode(basic_auth_payload).decode('ascii')}" }, ) assert utils.get_deposits_list() == deposits_data["results"] @pytest.mark.parametrize("backend", ["swh-search", "swh-storage"]) def test_origin_visit_types(mocker, backend): if backend != "swh-search": # equivalent to not configuring search in the config search = mocker.patch("swh.web.common.utils.search") search.return_value = None assert utils.origin_visit_types() == [] else: # see swh/web/tests/data.py for origins added for tests assert utils.origin_visit_types() == ["git", "tar"] @pytest.mark.parametrize("server_name", ["localhost", "127.0.0.1", "testserver"]) def test_is_swh_web_development(request_factory, server_name): request = request_factory.get("/", SERVER_NAME=server_name) assert utils.is_swh_web_development(request) @pytest.mark.parametrize("server_name", SWH_WEB_STAGING_SERVER_NAMES) def test_is_swh_web_staging(request_factory, server_name): request = request_factory.get("/", SERVER_NAME=server_name) assert utils.is_swh_web_staging(request) def test_is_swh_web_production(request_factory): request = request_factory.get("/", SERVER_NAME=SWH_WEB_SERVER_NAME) assert utils.is_swh_web_production(request) def add(x, y): return x + y def test_django_cache(mocker): """Decorated function should be called once and returned value put in django cache.""" spy_add = mocker.spy(sys.modules[__name__], "add") spy_cache_set = mocker.spy(utils.cache, "set") cached_add = utils.django_cache()(add) val = cached_add(1, 2) val2 = cached_add(1, 2) assert val == val2 == 3 assert spy_add.call_count == 1 assert spy_cache_set.call_count == 1 def test_django_cache_invalidate_cache_pred(mocker): """Decorated function should be called twice and returned value put in django cache twice.""" spy_add = mocker.spy(sys.modules[__name__], "add") spy_cache_set = mocker.spy(utils.cache, "set") cached_add = utils.django_cache(invalidate_cache_pred=lambda val: val == 3)(add) val = cached_add(1, 2) val2 = cached_add(1, 2) assert val == val2 == 3 assert spy_add.call_count == 2 assert spy_cache_set.call_count == 2 def test_django_cache_raise_exception(mocker): """Decorated function should be called twice, exceptions should be raised and no value put in django cache""" spy_add = mocker.spy(sys.modules[__name__], "add") spy_cache_set = mocker.spy(utils.cache, "set") cached_add = utils.django_cache()(add) with pytest.raises(TypeError): cached_add(1, "2") with pytest.raises(TypeError): cached_add(1, "2") assert spy_add.call_count == 2 assert spy_cache_set.call_count == 0 def test_django_cache_catch_exception(mocker): """Decorated function should be called twice, exceptions should not be raised, specified fallback value should be returned and no value put in django cache""" spy_add = mocker.spy(sys.modules[__name__], "add") spy_cache_set = mocker.spy(utils.cache, "set") cached_add = utils.django_cache( catch_exception=True, exception_return_value=math.nan )(add) val = cached_add(1, "2") val2 = cached_add(1, "2") assert math.isnan(val) assert math.isnan(val2) assert spy_add.call_count == 2 assert spy_cache_set.call_count == 0 diff --git a/swh/web/tests/conftest.py b/swh/web/tests/conftest.py index 09d087f8..f8311c25 100644 --- a/swh/web/tests/conftest.py +++ b/swh/web/tests/conftest.py @@ -1,1250 +1,1204 @@ # Copyright (C) 2018-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from collections import defaultdict from datetime import timedelta import functools import json import os import random import shutil -from subprocess import PIPE, run import sys import time from typing import Any, Dict, List, Optional from _pytest.python import Function from hypothesis import HealthCheck, settings import pytest from django.contrib.auth.models import User from django.core.cache import cache from django.test.utils import setup_databases from rest_framework.test import APIClient, APIRequestFactory from swh.model.hashutil import ( ALGORITHMS, DEFAULT_ALGORITHMS, hash_to_bytes, hash_to_hex, ) from swh.model.model import Content, Directory from swh.model.swhids import CoreSWHID, ObjectType from swh.scheduler.tests.common import TASK_TYPES from swh.storage.algos.origin import origin_get_latest_visit_status from swh.storage.algos.revisions_walker import get_revisions_walker from swh.storage.algos.snapshot import snapshot_get_all_branches, snapshot_get_latest from swh.web.auth.utils import ( ADD_FORGE_MODERATOR_PERMISSION, MAILMAP_ADMIN_PERMISSION, MAILMAP_PERMISSION, OIDC_SWH_WEB_CLIENT_ID, ) from swh.web.common import converters from swh.web.common.origin_save import get_scheduler_load_task_types from swh.web.common.typing import OriginVisitInfo from swh.web.common.utils import browsers_supported_image_mimes from swh.web.config import get_config from swh.web.tests.data import ( get_tests_data, override_storages, random_content, random_sha1, random_sha1_bytes, random_sha256, ) from swh.web.tests.utils import create_django_permission os.environ["LC_ALL"] = "C.UTF-8" -# Used to skip some tests -ctags_json_missing = ( - shutil.which("ctags") is None - or b"+json" not in run(["ctags", "--version"], stdout=PIPE).stdout -) - fossology_missing = shutil.which("nomossa") is None # Register some hypothesis profiles settings.register_profile("default", settings()) # we use getattr here to keep mypy happy regardless hypothesis version function_scoped_fixture_check = ( [getattr(HealthCheck, "function_scoped_fixture")] if hasattr(HealthCheck, "function_scoped_fixture") else [] ) suppress_health_check = [ HealthCheck.too_slow, HealthCheck.filter_too_much, ] + function_scoped_fixture_check settings.register_profile( "swh-web", settings( deadline=None, suppress_health_check=suppress_health_check, ), ) settings.register_profile( "swh-web-fast", settings( deadline=None, max_examples=5, suppress_health_check=suppress_health_check, ), ) def pytest_addoption(parser): parser.addoption("--swh-web-random-seed", action="store", default=None) def pytest_configure(config): # Use fast hypothesis profile by default if none has been # explicitly specified in pytest option if config.getoption("--hypothesis-profile") is None: settings.load_profile("swh-web-fast") # Small hack in order to be able to run the unit tests # without static assets generated by webpack. # Those assets are not really needed for the Python tests # but the django templates will fail to load due to missing # generated file webpack-stats.json describing the js and css # files to include. # So generate a dummy webpack-stats.json file to overcome # that issue. test_dir = os.path.dirname(__file__) # location of the static folder when running tests through tox data_dir = os.path.join(sys.prefix, "share/swh/web") static_dir = os.path.join(data_dir, "static") if not os.path.exists(static_dir): # location of the static folder when running tests locally with pytest static_dir = os.path.join(test_dir, "../../../static") webpack_stats = os.path.join(static_dir, "webpack-stats.json") if os.path.exists(webpack_stats): return bundles_dir = os.path.join(test_dir, "../../../assets/src/bundles") if not os.path.exists(bundles_dir): # location of the bundles folder when running tests with tox bundles_dir = os.path.join(data_dir, "assets/src/bundles") _, bundles, _ = next(os.walk(bundles_dir)) mock_webpack_stats = { "status": "done", "publicPath": "/static", "chunks": {}, "assets": {}, } for bundle in bundles: asset = f"js/{bundle}.js" mock_webpack_stats["chunks"][bundle] = [asset] mock_webpack_stats["assets"][asset] = { "name": asset, "publicPath": f"/static/{asset}", } with open(webpack_stats, "w") as outfile: json.dump(mock_webpack_stats, outfile) _swh_web_custom_section = "swh-web custom section" _random_seed_cache_key = "swh-web/random-seed" @pytest.fixture(scope="function", autouse=True) def random_seed(pytestconfig): state = random.getstate() seed = pytestconfig.getoption("--swh-web-random-seed") if seed is None: seed = time.time() seed = int(seed) cache.set(_random_seed_cache_key, seed) random.seed(seed) yield seed random.setstate(state) def pytest_report_teststatus(report, *args): if report.when == "call" and report.outcome == "failed": seed = cache.get(_random_seed_cache_key, None) line = ( f'FAILED {report.nodeid}: Use "pytest --swh-web-random-seed={seed} ' f'{report.nodeid}" to reproduce that test failure with same inputs' ) report.sections.append((_swh_web_custom_section, line)) def pytest_terminal_summary(terminalreporter, *args): reports = terminalreporter.getreports("failed") content = os.linesep.join( text for report in reports for secname, text in report.sections if secname == _swh_web_custom_section ) if content: terminalreporter.ensure_newline() terminalreporter.section(_swh_web_custom_section, sep="-", blue=True, bold=True) terminalreporter.line(content) # Clear Django cache before each test @pytest.fixture(autouse=True) def django_cache_cleared(): cache.clear() # Alias rf fixture from pytest-django @pytest.fixture def request_factory(rf): return rf # Fixture to get test client from Django REST Framework @pytest.fixture def api_client(): return APIClient() # Fixture to get API request factory from Django REST Framework @pytest.fixture def api_request_factory(): return APIRequestFactory() # Initialize tests data @pytest.fixture(scope="function", autouse=True) def tests_data(): data = get_tests_data(reset=True) # Update swh-web configuration to use the in-memory storages # instantiated in the tests.data module override_storages( data["storage"], data["idx_storage"], data["search"], data["counters"] ) return data @pytest.fixture(scope="function") def sha1(): """Fixture returning a valid hexadecimal sha1 value.""" return random_sha1() @pytest.fixture(scope="function") def invalid_sha1(): """Fixture returning an invalid sha1 representation.""" return hash_to_hex(bytes(random.randint(0, 255) for _ in range(50))) @pytest.fixture(scope="function") def sha256(): """Fixture returning a valid hexadecimal sha256 value.""" return random_sha256() def _known_swh_objects(tests_data, object_type): return tests_data[object_type] @pytest.fixture(scope="function") def content(tests_data): """Fixture returning a random content ingested into the test archive.""" return random.choice(_known_swh_objects(tests_data, "contents")) @pytest.fixture(scope="function") def contents(tests_data): """Fixture returning random contents ingested into the test archive.""" return random.choices( _known_swh_objects(tests_data, "contents"), k=random.randint(2, 8) ) def _new_content(tests_data): while True: new_content = random_content() sha1_bytes = hash_to_bytes(new_content["sha1"]) if tests_data["storage"].content_get_data(sha1_bytes) is None: return new_content @pytest.fixture(scope="function") def unknown_content(tests_data): """Fixture returning a random content not ingested into the test archive.""" return _new_content(tests_data) @pytest.fixture(scope="function") def unknown_contents(tests_data): """Fixture returning random contents not ingested into the test archive.""" new_contents = [] new_content_ids = set() nb_contents = random.randint(2, 8) while len(new_contents) != nb_contents: new_content = _new_content(tests_data) if new_content["sha1"] not in new_content_ids: new_contents.append(new_content) new_content_ids.add(new_content["sha1"]) return list(new_contents) @pytest.fixture(scope="function") def empty_content(): """Fixture returning the empty content ingested into the test archive.""" empty_content = Content.from_data(data=b"").to_dict() for algo in DEFAULT_ALGORITHMS: empty_content[algo] = hash_to_hex(empty_content[algo]) return empty_content @functools.lru_cache(maxsize=None) def _content_text(): return list( filter( lambda c: c["mimetype"].startswith("text/"), _known_swh_objects(get_tests_data(), "contents"), ) ) @pytest.fixture(scope="function") def content_text(): """ Fixture returning a random textual content ingested into the test archive. """ return random.choice(_content_text()) @functools.lru_cache(maxsize=None) def _content_text_non_utf8(): return list( filter( lambda c: c["mimetype"].startswith("text/") and c["encoding"] not in ("utf-8", "us-ascii"), _known_swh_objects(get_tests_data(), "contents"), ) ) @pytest.fixture(scope="function") def content_text_non_utf8(): """Fixture returning a random textual content not encoded to UTF-8 ingested into the test archive. """ return random.choice(_content_text_non_utf8()) @functools.lru_cache(maxsize=None) def _content_application_no_highlight(): return list( filter( lambda c: c["mimetype"].startswith("application/") and c["hljs_language"] == "plaintext", _known_swh_objects(get_tests_data(), "contents"), ) ) @pytest.fixture(scope="function") def content_application_no_highlight(): """Fixture returning a random textual content with mimetype starting with application/ and no detected programming language to highlight ingested into the test archive. """ return random.choice(_content_application_no_highlight()) @functools.lru_cache(maxsize=None) def _content_text_no_highlight(): return list( filter( lambda c: c["mimetype"].startswith("text/") and c["hljs_language"] == "plaintext", _known_swh_objects(get_tests_data(), "contents"), ) ) @pytest.fixture(scope="function") def content_text_no_highlight(): """Fixture returning a random textual content with no detected programming language to highlight ingested into the test archive. """ return random.choice(_content_text_no_highlight()) @functools.lru_cache(maxsize=None) def _content_image_type(): return list( filter( lambda c: c["mimetype"] in browsers_supported_image_mimes, _known_swh_objects(get_tests_data(), "contents"), ) ) @pytest.fixture(scope="function") def content_image_type(): """Fixture returning a random image content ingested into the test archive.""" return random.choice(_content_image_type()) @functools.lru_cache(maxsize=None) def _content_unsupported_image_type_rendering(): return list( filter( lambda c: c["mimetype"].startswith("image/") and c["mimetype"] not in browsers_supported_image_mimes, _known_swh_objects(get_tests_data(), "contents"), ) ) @pytest.fixture(scope="function") def content_unsupported_image_type_rendering(): """Fixture returning a random image content ingested into the test archive that can not be rendered by browsers. """ return random.choice(_content_unsupported_image_type_rendering()) @functools.lru_cache(maxsize=None) def _content_utf8_detected_as_binary(): def utf8_binary_detected(content): if content["encoding"] != "binary": return False try: content["raw_data"].decode("utf-8") except Exception: return False else: return True return list( filter(utf8_binary_detected, _known_swh_objects(get_tests_data(), "contents")) ) @pytest.fixture(scope="function") def content_utf8_detected_as_binary(): """Fixture returning a random textual content detected as binary by libmagic while they are valid UTF-8 encoded files. """ return random.choice(_content_utf8_detected_as_binary()) -@pytest.fixture(scope="function") -def contents_with_ctags(): - """ - Fixture returning contents ingested into the test archive. - Those contents are ctags compatible, that is running ctags on those lay results. - """ - return { - "sha1s": [ - "0ab37c02043ebff946c1937523f60aadd0844351", - "15554cf7608dde6bfefac7e3d525596343a85b6f", - "2ce837f1489bdfb8faf3ebcc7e72421b5bea83bd", - "30acd0b47fc25e159e27a980102ddb1c4bea0b95", - "4f81f05aaea3efb981f9d90144f746d6b682285b", - "5153aa4b6e4455a62525bc4de38ed0ff6e7dd682", - "59d08bafa6a749110dfb65ba43a61963d5a5bf9f", - "7568285b2d7f31ae483ae71617bd3db873deaa2c", - "7ed3ee8e94ac52ba983dd7690bdc9ab7618247b4", - "8ed7ef2e7ff9ed845e10259d08e4145f1b3b5b03", - "9b3557f1ab4111c8607a4f2ea3c1e53c6992916c", - "9c20da07ed14dc4fcd3ca2b055af99b2598d8bdd", - "c20ceebd6ec6f7a19b5c3aebc512a12fbdc9234b", - "e89e55a12def4cd54d5bff58378a3b5119878eb7", - "e8c0654fe2d75ecd7e0b01bee8a8fc60a130097e", - "eb6595e559a1d34a2b41e8d4835e0e4f98a5d2b5", - ], - "symbol_name": "ABS", - } - - @pytest.fixture(scope="function") def directory(tests_data): """Fixture returning a random directory ingested into the test archive.""" return random.choice(_known_swh_objects(tests_data, "directories")) @functools.lru_cache(maxsize=None) def _directory_with_entry_type(type_): tests_data = get_tests_data() return list( filter( lambda d: any( [ e["type"] == type_ for e in list(tests_data["storage"].directory_ls(hash_to_bytes(d))) ] ), _known_swh_objects(tests_data, "directories"), ) ) @pytest.fixture(scope="function") def directory_with_subdirs(): """Fixture returning a random directory containing sub directories ingested into the test archive. """ return random.choice(_directory_with_entry_type("dir")) @pytest.fixture(scope="function") def directory_with_files(): """Fixture returning a random directory containing at least one regular file.""" return random.choice(_directory_with_entry_type("file")) @pytest.fixture(scope="function") def unknown_directory(tests_data): """Fixture returning a random directory not ingested into the test archive.""" while True: new_directory = random_sha1() sha1_bytes = hash_to_bytes(new_directory) if list(tests_data["storage"].directory_missing([sha1_bytes])): return new_directory @pytest.fixture(scope="function") def empty_directory(): """Fixture returning the empty directory ingested into the test archive.""" return Directory(entries=()).id.hex() @pytest.fixture(scope="function") def revision(tests_data): """Fixturereturning a random revision ingested into the test archive.""" return random.choice(_known_swh_objects(tests_data, "revisions")) @pytest.fixture(scope="function") def revisions(tests_data): """Fixture returning random revisions ingested into the test archive.""" return random.choices( _known_swh_objects(tests_data, "revisions"), k=random.randint(2, 8), ) @pytest.fixture(scope="function") def revisions_list(tests_data): """Fixture returning random revisions ingested into the test archive.""" def gen_revisions_list(size): return random.choices( _known_swh_objects(tests_data, "revisions"), k=size, ) return gen_revisions_list @pytest.fixture(scope="function") def unknown_revision(tests_data): """Fixture returning a random revision not ingested into the test archive.""" while True: new_revision = random_sha1() sha1_bytes = hash_to_bytes(new_revision) if tests_data["storage"].revision_get([sha1_bytes])[0] is None: return new_revision def _get_origin_dfs_revisions_walker(tests_data): storage = tests_data["storage"] origin = random.choice(tests_data["origins"][:-1]) snapshot = snapshot_get_latest(storage, origin["url"]) if snapshot.branches[b"HEAD"].target_type.value == "alias": target = snapshot.branches[b"HEAD"].target head = snapshot.branches[target].target else: head = snapshot.branches[b"HEAD"].target return get_revisions_walker("dfs", storage, head) @functools.lru_cache(maxsize=None) def _ancestor_revisions_data(): # get a dfs revisions walker for one of the origins # loaded into the test archive revisions_walker = _get_origin_dfs_revisions_walker(get_tests_data()) master_revisions = [] children = defaultdict(list) init_rev_found = False # get revisions only authored in the master branch for rev in revisions_walker: for rev_p in rev["parents"]: children[rev_p].append(rev["id"]) if not init_rev_found: master_revisions.append(rev) if not rev["parents"]: init_rev_found = True return master_revisions, children @pytest.fixture(scope="function") def ancestor_revisions(): """Fixture returning a pair of revisions ingested into the test archive with an ancestor relation. """ master_revisions, children = _ancestor_revisions_data() # head revision root_rev = master_revisions[0] # pick a random revision, different from head, only authored # in the master branch ancestor_rev_idx = random.choice(list(range(1, len(master_revisions) - 1))) ancestor_rev = master_revisions[ancestor_rev_idx] ancestor_child_revs = children[ancestor_rev["id"]] return { "sha1_git_root": hash_to_hex(root_rev["id"]), "sha1_git": hash_to_hex(ancestor_rev["id"]), "children": [hash_to_hex(r) for r in ancestor_child_revs], } @functools.lru_cache(maxsize=None) def _non_ancestor_revisions_data(): # get a dfs revisions walker for one of the origins # loaded into the test archive revisions_walker = _get_origin_dfs_revisions_walker(get_tests_data()) merge_revs = [] children = defaultdict(list) # get all merge revisions for rev in revisions_walker: if len(rev["parents"]) > 1: merge_revs.append(rev) for rev_p in rev["parents"]: children[rev_p].append(rev["id"]) return merge_revs, children @pytest.fixture(scope="function") def non_ancestor_revisions(): """Fixture returning a pair of revisions ingested into the test archive with no ancestor relation. """ merge_revs, children = _non_ancestor_revisions_data() # find a merge revisions whose parents have a unique child revision random.shuffle(merge_revs) selected_revs = None for merge_rev in merge_revs: if all(len(children[rev_p]) == 1 for rev_p in merge_rev["parents"]): selected_revs = merge_rev["parents"] return { "sha1_git_root": hash_to_hex(selected_revs[0]), "sha1_git": hash_to_hex(selected_revs[1]), } @pytest.fixture(scope="function") def revision_with_submodules(): """Fixture returning a revision that is known to point to a directory with revision entries (aka git submodules) """ return { "rev_sha1_git": "ffcb69001f3f6745dfd5b48f72ab6addb560e234", "rev_dir_sha1_git": "d92a21446387fa28410e5a74379c934298f39ae2", "rev_dir_rev_path": "libtess2", } @pytest.fixture(scope="function") def release(tests_data): """Fixture returning a random release ingested into the test archive.""" return random.choice(_known_swh_objects(tests_data, "releases")) @pytest.fixture(scope="function") def releases(tests_data): """Fixture returning random releases ingested into the test archive.""" return random.choices( _known_swh_objects(tests_data, "releases"), k=random.randint(2, 8) ) @pytest.fixture(scope="function") def unknown_release(tests_data): """Fixture returning a random release not ingested into the test archive.""" while True: new_release = random_sha1() sha1_bytes = hash_to_bytes(new_release) if tests_data["storage"].release_get([sha1_bytes])[0] is None: return new_release @pytest.fixture(scope="function") def snapshot(tests_data): """Fixture returning a random snapshot ingested into the test archive.""" return random.choice(_known_swh_objects(tests_data, "snapshots")) @pytest.fixture(scope="function") def unknown_snapshot(tests_data): """Fixture returning a random snapshot not ingested into the test archive.""" while True: new_snapshot = random_sha1() sha1_bytes = hash_to_bytes(new_snapshot) if tests_data["storage"].snapshot_get_branches(sha1_bytes) is None: return new_snapshot @pytest.fixture(scope="function") def origin(tests_data): """Fixture returning a random origin ingested into the test archive.""" return random.choice(_known_swh_objects(tests_data, "origins")) @functools.lru_cache(maxsize=None) def _origin_with_multiple_visits(): tests_data = get_tests_data() origins = [] storage = tests_data["storage"] for origin in tests_data["origins"]: visit_page = storage.origin_visit_get(origin["url"]) if len(visit_page.results) > 1: origins.append(origin) return origins @pytest.fixture(scope="function") def origin_with_multiple_visits(): """Fixture returning a random origin with multiple visits ingested into the test archive. """ return random.choice(_origin_with_multiple_visits()) @functools.lru_cache(maxsize=None) def _origin_with_releases(): tests_data = get_tests_data() origins = [] for origin in tests_data["origins"]: snapshot = snapshot_get_latest(tests_data["storage"], origin["url"]) if any([b.target_type.value == "release" for b in snapshot.branches.values()]): origins.append(origin) return origins @pytest.fixture(scope="function") def origin_with_releases(): """Fixture returning a random origin with releases ingested into the test archive.""" return random.choice(_origin_with_releases()) @functools.lru_cache(maxsize=None) def _origin_with_pull_request_branches(): tests_data = get_tests_data() origins = [] storage = tests_data["storage"] for origin in storage.origin_list(limit=1000).results: snapshot = snapshot_get_latest(storage, origin.url) if any([b"refs/pull/" in b for b in snapshot.branches]): origins.append(origin) return origins @pytest.fixture(scope="function") def origin_with_pull_request_branches(): """Fixture returning a random origin with pull request branches ingested into the test archive. """ return random.choice(_origin_with_pull_request_branches()) @functools.lru_cache(maxsize=None) def _object_type_swhid(object_type): return list( filter( lambda swhid: swhid.object_type == object_type, _known_swh_objects(get_tests_data(), "swhids"), ) ) @pytest.fixture(scope="function") def content_swhid(): """Fixture returning a qualified SWHID for a random content object ingested into the test archive. """ return random.choice(_object_type_swhid(ObjectType.CONTENT)) @pytest.fixture(scope="function") def directory_swhid(): """Fixture returning a qualified SWHID for a random directory object ingested into the test archive. """ return random.choice(_object_type_swhid(ObjectType.DIRECTORY)) @pytest.fixture(scope="function") def release_swhid(): """Fixture returning a qualified SWHID for a random release object ingested into the test archive. """ return random.choice(_object_type_swhid(ObjectType.RELEASE)) @pytest.fixture(scope="function") def revision_swhid(): """Fixture returning a qualified SWHID for a random revision object ingested into the test archive. """ return random.choice(_object_type_swhid(ObjectType.REVISION)) @pytest.fixture(scope="function") def snapshot_swhid(): """Fixture returning a qualified SWHID for a snapshot object ingested into the test archive. """ return random.choice(_object_type_swhid(ObjectType.SNAPSHOT)) @pytest.fixture(scope="function", params=list(ObjectType)) def unknown_core_swhid(request) -> CoreSWHID: """Fixture returning an unknown core SWHID. Tests using this will be called once per object type. """ return CoreSWHID( object_type=request.param, object_id=random_sha1_bytes(), ) # Fixture to manipulate data from a sample archive used in the tests @pytest.fixture(scope="function") def archive_data(tests_data): return _ArchiveData(tests_data) # Fixture to manipulate indexer data from a sample archive used in the tests @pytest.fixture(scope="function") def indexer_data(tests_data): return _IndexerData(tests_data) # Custom data directory for requests_mock @pytest.fixture def datadir(): return os.path.join(os.path.abspath(os.path.dirname(__file__)), "resources") class _ArchiveData: """ Helper class to manage data from a sample test archive. It is initialized with a reference to an in-memory storage containing raw tests data. It is basically a proxy to Storage interface but it overrides some methods to retrieve those tests data in a json serializable format in order to ease tests implementation. """ def __init__(self, tests_data): self.storage = tests_data["storage"] def __getattr__(self, key): if key == "storage": raise AttributeError(key) # Forward calls to non overridden Storage methods to wrapped # storage instance return getattr(self.storage, key) def content_find(self, content: Dict[str, Any]) -> Dict[str, Any]: cnt_ids_bytes = { algo_hash: hash_to_bytes(content[algo_hash]) for algo_hash in ALGORITHMS if content.get(algo_hash) } cnt = self.storage.content_find(cnt_ids_bytes) return converters.from_content(cnt[0].to_dict()) if cnt else cnt def content_get(self, cnt_id: str) -> Dict[str, Any]: cnt_id_bytes = hash_to_bytes(cnt_id) content = self.storage.content_get([cnt_id_bytes])[0] if content: content_d = content.to_dict() content_d.pop("ctime", None) else: content_d = None return converters.from_swh( content_d, hashess={"sha1", "sha1_git", "sha256", "blake2s256"} ) def content_get_data(self, cnt_id: str) -> Optional[Dict[str, Any]]: cnt_id_bytes = hash_to_bytes(cnt_id) cnt_data = self.storage.content_get_data(cnt_id_bytes) if cnt_data is None: return None return converters.from_content({"data": cnt_data, "sha1": cnt_id_bytes}) def directory_get(self, dir_id): return {"id": dir_id, "content": self.directory_ls(dir_id)} def directory_ls(self, dir_id): cnt_id_bytes = hash_to_bytes(dir_id) dir_content = map( converters.from_directory_entry, self.storage.directory_ls(cnt_id_bytes) ) return list(dir_content) def release_get(self, rel_id: str) -> Optional[Dict[str, Any]]: rel_id_bytes = hash_to_bytes(rel_id) rel_data = self.storage.release_get([rel_id_bytes])[0] return converters.from_release(rel_data) if rel_data else None def revision_get(self, rev_id: str) -> Optional[Dict[str, Any]]: rev_id_bytes = hash_to_bytes(rev_id) rev_data = self.storage.revision_get([rev_id_bytes])[0] return converters.from_revision(rev_data) if rev_data else None def revision_log(self, rev_id, limit=None): rev_id_bytes = hash_to_bytes(rev_id) return list( map( converters.from_revision, self.storage.revision_log([rev_id_bytes], limit=limit), ) ) def snapshot_get_latest(self, origin_url): snp = snapshot_get_latest(self.storage, origin_url) return converters.from_snapshot(snp.to_dict()) def origin_get(self, origin_urls): origins = self.storage.origin_get(origin_urls) return [converters.from_origin(o.to_dict()) for o in origins] def origin_visit_get(self, origin_url): next_page_token = None visits = [] while True: visit_page = self.storage.origin_visit_get( origin_url, page_token=next_page_token ) next_page_token = visit_page.next_page_token for visit in visit_page.results: visit_status = self.storage.origin_visit_status_get_latest( origin_url, visit.visit ) visits.append( converters.from_origin_visit( {**visit_status.to_dict(), "type": visit.type} ) ) if not next_page_token: break return visits def origin_visit_get_by(self, origin_url: str, visit_id: int) -> OriginVisitInfo: visit = self.storage.origin_visit_get_by(origin_url, visit_id) assert visit is not None visit_status = self.storage.origin_visit_status_get_latest(origin_url, visit_id) assert visit_status is not None return converters.from_origin_visit( {**visit_status.to_dict(), "type": visit.type} ) def origin_visit_status_get_latest( self, origin_url, type: Optional[str] = None, allowed_statuses: Optional[List[str]] = None, require_snapshot: bool = False, ): visit_status = origin_get_latest_visit_status( self.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 snapshot_get(self, snapshot_id): snp = snapshot_get_all_branches(self.storage, hash_to_bytes(snapshot_id)) return converters.from_snapshot(snp.to_dict()) def snapshot_get_branches( self, snapshot_id, branches_from="", branches_count=1000, target_types=None ): partial_branches = self.storage.snapshot_get_branches( hash_to_bytes(snapshot_id), branches_from.encode(), branches_count, target_types, ) return converters.from_partial_branches(partial_branches) def snapshot_get_head(self, snapshot): if snapshot["branches"]["HEAD"]["target_type"] == "alias": target = snapshot["branches"]["HEAD"]["target"] head = snapshot["branches"][target]["target"] else: head = snapshot["branches"]["HEAD"]["target"] return head def snapshot_count_branches(self, snapshot_id): counts = dict.fromkeys(("alias", "release", "revision"), 0) counts.update(self.storage.snapshot_count_branches(hash_to_bytes(snapshot_id))) counts.pop(None, None) return counts class _IndexerData: """ Helper class to manage indexer tests data It is initialized with a reference to an in-memory indexer storage containing raw tests data. It also defines class methods to retrieve those tests data in a json serializable format in order to ease tests implementation. """ def __init__(self, tests_data): self.idx_storage = tests_data["idx_storage"] self.mimetype_indexer = tests_data["mimetype_indexer"] self.license_indexer = tests_data["license_indexer"] - self.ctags_indexer = tests_data["ctags_indexer"] def content_add_mimetype(self, cnt_id): self.mimetype_indexer.run([hash_to_bytes(cnt_id)]) def content_get_mimetype(self, cnt_id): mimetype = self.idx_storage.content_mimetype_get([hash_to_bytes(cnt_id)])[ 0 ].to_dict() return converters.from_filetype(mimetype) def content_add_license(self, cnt_id): self.license_indexer.run([hash_to_bytes(cnt_id)]) def content_get_license(self, cnt_id): cnt_id_bytes = hash_to_bytes(cnt_id) licenses = self.idx_storage.content_fossology_license_get([cnt_id_bytes]) for license in licenses: yield converters.from_swh(license.to_dict(), hashess={"id"}) - def content_add_ctags(self, cnt_id): - self.ctags_indexer.run([hash_to_bytes(cnt_id)]) - - def content_get_ctags(self, cnt_id): - cnt_id_bytes = hash_to_bytes(cnt_id) - ctags = self.idx_storage.content_ctags_get([cnt_id_bytes]) - for ctag in ctags: - yield converters.from_swh(ctag, hashess={"id"}) - @pytest.fixture def keycloak_oidc(keycloak_oidc, mocker): keycloak_config = get_config()["keycloak"] keycloak_oidc.server_url = keycloak_config["server_url"] keycloak_oidc.realm_name = keycloak_config["realm_name"] keycloak_oidc.client_id = OIDC_SWH_WEB_CLIENT_ID keycloak_oidc_client = mocker.patch("swh.web.auth.views.keycloak_oidc_client") keycloak_oidc_client.return_value = keycloak_oidc return keycloak_oidc @pytest.fixture def subtest(request): """A hack to explicitly set up and tear down fixtures. This fixture allows you to set up and tear down fixtures within the test function itself. This is useful (necessary!) for using Hypothesis inside pytest, as hypothesis will call the test function multiple times, without setting up or tearing down fixture state as it is normally the case. Copied from the pytest-subtesthack project, public domain license (https://github.com/untitaker/pytest-subtesthack). """ parent_test = request.node def inner(func): if hasattr(Function, "from_parent"): item = Function.from_parent( parent_test, name=request.function.__name__ + "[]", originalname=request.function.__name__, callobj=func, ) else: item = Function( name=request.function.__name__ + "[]", parent=parent_test, callobj=func ) nextitem = parent_test # prevents pytest from tearing down module fixtures item.ihook.pytest_runtest_setup(item=item) try: item.ihook.pytest_runtest_call(item=item) finally: item.ihook.pytest_runtest_teardown(item=item, nextitem=nextitem) return inner @pytest.fixture def swh_scheduler(swh_scheduler): config = get_config() scheduler = config["scheduler"] config["scheduler"] = swh_scheduler # create load-git and load-hg task types for task_type in TASK_TYPES.values(): # see https://forge.softwareheritage.org/rDSCHc46ffadf7adf24c7eb3ffce062e8ade3818c79cc # noqa task_type["type"] = task_type["type"].replace("load-test-", "load-", 1) swh_scheduler.create_task_type(task_type) # create load-svn task type swh_scheduler.create_task_type( { "type": "load-svn", "description": "Update a Subversion repository", "backend_name": "swh.loader.svn.tasks.DumpMountAndLoadSvnRepository", "default_interval": timedelta(days=64), "min_interval": timedelta(hours=12), "max_interval": timedelta(days=64), "backoff_factor": 2, "max_queue_length": None, "num_retries": 7, "retry_delay": timedelta(hours=2), } ) # create load-cvs task type swh_scheduler.create_task_type( { "type": "load-cvs", "description": "Update a CVS repository", "backend_name": "swh.loader.cvs.tasks.DumpMountAndLoadSvnRepository", "default_interval": timedelta(days=64), "min_interval": timedelta(hours=12), "max_interval": timedelta(days=64), "backoff_factor": 2, "max_queue_length": None, "num_retries": 7, "retry_delay": timedelta(hours=2), } ) # create load-bzr task type swh_scheduler.create_task_type( { "type": "load-bzr", "description": "Update a Bazaar repository", "backend_name": "swh.loader.bzr.tasks.LoadBazaar", "default_interval": timedelta(days=64), "min_interval": timedelta(hours=12), "max_interval": timedelta(days=64), "backoff_factor": 2, "max_queue_length": None, "num_retries": 7, "retry_delay": timedelta(hours=2), } ) # add method to add load-archive-files task type during tests def add_load_archive_task_type(): swh_scheduler.create_task_type( { "type": "load-archive-files", "description": "Load tarballs", "backend_name": "swh.loader.package.archive.tasks.LoadArchive", "default_interval": timedelta(days=64), "min_interval": timedelta(hours=12), "max_interval": timedelta(days=64), "backoff_factor": 2, "max_queue_length": None, "num_retries": 7, "retry_delay": timedelta(hours=2), } ) swh_scheduler.add_load_archive_task_type = add_load_archive_task_type yield swh_scheduler config["scheduler"] = scheduler get_scheduler_load_task_types.cache_clear() @pytest.fixture(scope="session") def django_db_setup(request, django_db_blocker, postgresql_proc): from django.conf import settings settings.DATABASES["default"].update( { ("ENGINE", "django.db.backends.postgresql"), ("NAME", get_config()["test_db"]["name"]), ("USER", postgresql_proc.user), ("HOST", postgresql_proc.host), ("PORT", postgresql_proc.port), } ) with django_db_blocker.unblock(): setup_databases( verbosity=request.config.option.verbose, interactive=False, keepdb=False ) @pytest.fixture def staff_user(): return User.objects.create_user(username="admin", password="", is_staff=True) @pytest.fixture def regular_user(): return User.objects.create_user(username="johndoe", password="") @pytest.fixture def regular_user2(): return User.objects.create_user(username="janedoe", password="") @pytest.fixture def add_forge_moderator(): moderator = User.objects.create_user(username="add-forge moderator", password="") moderator.user_permissions.add( create_django_permission(ADD_FORGE_MODERATOR_PERMISSION) ) return moderator @pytest.fixture def mailmap_admin(): mailmap_admin = User.objects.create_user(username="mailmap-admin", password="") mailmap_admin.user_permissions.add( create_django_permission(MAILMAP_ADMIN_PERMISSION) ) return mailmap_admin @pytest.fixture def mailmap_user(): mailmap_user = User.objects.create_user(username="mailmap-user", password="") mailmap_user.user_permissions.add(create_django_permission(MAILMAP_PERMISSION)) return mailmap_user diff --git a/swh/web/tests/data.py b/swh/web/tests/data.py index f81744bf..d6cee83c 100644 --- a/swh/web/tests/data.py +++ b/swh/web/tests/data.py @@ -1,561 +1,542 @@ -# Copyright (C) 2018-2021 The Software Heritage developers +# Copyright (C) 2018-2022 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from copy import deepcopy from datetime import timedelta import os from pathlib import Path import random import time from typing import Dict, List, Optional, Set from swh.core.config import merge_configs from swh.counters import get_counters -from swh.indexer.ctags import CtagsIndexer from swh.indexer.fossology_license import FossologyLicenseIndexer from swh.indexer.mimetype import MimetypeIndexer from swh.indexer.storage import get_indexer_storage from swh.indexer.storage.model import OriginIntrinsicMetadataRow from swh.loader.git.from_disk import GitLoaderFromArchive from swh.model.hashutil import DEFAULT_ALGORITHMS, hash_to_hex from swh.model.model import ( Content, Directory, Origin, OriginVisit, OriginVisitStatus, Snapshot, ) from swh.model.swhids import CoreSWHID, ObjectType, QualifiedSWHID from swh.search import get_search from swh.storage import get_storage from swh.storage.algos.dir_iterators import dir_iterator from swh.storage.algos.snapshot import snapshot_get_latest from swh.storage.interface import Sha1 from swh.storage.utils import now from swh.web import config from swh.web.browse.utils import ( get_mimetype_and_encoding_for_content, prepare_content_for_display, re_encode_content, ) from swh.web.common import archive # Module used to initialize data that will be provided as tests input # Base content indexer configuration _TEST_INDEXER_BASE_CONFIG = { "storage": {"cls": "memory"}, "objstorage": { "cls": "memory", "args": {}, }, "indexer_storage": { "cls": "memory", "args": {}, }, } def random_sha1_bytes() -> Sha1: return bytes(random.randint(0, 255) for _ in range(20)) def random_sha1() -> str: return hash_to_hex(random_sha1_bytes()) def random_sha256() -> str: return hash_to_hex(bytes(random.randint(0, 255) for _ in range(32))) def random_blake2s256() -> str: return hash_to_hex(bytes(random.randint(0, 255) for _ in range(32))) def random_content(): return { "sha1": random_sha1(), "sha1_git": random_sha1(), "sha256": random_sha256(), "blake2s256": random_blake2s256(), } _TEST_MIMETYPE_INDEXER_CONFIG = merge_configs( _TEST_INDEXER_BASE_CONFIG, { "tools": { "name": "file", "version": "1:5.30-1+deb9u1", "configuration": {"type": "library", "debian-package": "python3-magic"}, } }, ) _TEST_LICENSE_INDEXER_CONFIG = merge_configs( _TEST_INDEXER_BASE_CONFIG, { "workdir": "/tmp/swh/indexer.fossology.license", "tools": { "name": "nomos", "version": "3.1.0rc2-31-ga2cbb8c", "configuration": { "command_line": "nomossa ", }, }, }, ) -_TEST_CTAGS_INDEXER_CONFIG = merge_configs( - _TEST_INDEXER_BASE_CONFIG, - { - "workdir": "/tmp/swh/indexer.ctags", - "languages": {"c": "c"}, - "tools": { - "name": "universal-ctags", - "version": "~git7859817b", - "configuration": { - "command_line": """ctags --fields=+lnz --sort=no --links=no """ - """--output-format=json """ - }, - }, - }, -) - - # Lightweight git repositories that will be loaded to generate # input data for tests _TEST_ORIGINS = [ { "type": "git", "url": "https://github.com/memononen/libtess2", "archives": ["libtess2.zip"], "metadata": { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "description": ( "Game and tools oriented refactored version of GLU tessellator." ), }, }, { "type": "git", "url": "https://github.com/wcoder/highlightjs-line-numbers.js", "archives": [ "highlightjs-line-numbers.js.zip", "highlightjs-line-numbers.js_visit2.zip", ], "metadata": { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "description": "Line numbering plugin for Highlight.js", }, }, { "type": "git", "url": "repo_with_submodules", "archives": ["repo_with_submodules.tgz"], "metadata": { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "description": "This is just a sample repository with submodules", }, }, ] _contents = {} def _add_extra_contents(storage, contents): pbm_image_data = b"""P1 # PBM example 24 7 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 1 1 1 1 0 0 1 1 1 1 0 0 1 1 1 1 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 1 0 0 1 1 1 0 0 0 1 1 1 0 0 0 1 1 1 0 0 0 1 1 1 1 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 1 1 1 0 0 1 1 1 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0""" # add file with mimetype image/x-portable-bitmap in the archive content pbm_content = Content.from_data(pbm_image_data) storage.content_add([pbm_content]) contents.add(pbm_content.sha1) # add file with mimetype application/pgp-keys in the archive content gpg_path = os.path.join( os.path.dirname(__file__), "resources/contents/other/extensions/public.gpg" ) gpg_content = Content.from_data(Path(gpg_path).read_bytes()) storage.content_add([gpg_content]) contents.add(gpg_content.sha1) INDEXER_TOOL = { "tool_name": "swh-web tests", "tool_version": "1.0", "tool_configuration": {}, } ORIGIN_METADATA_KEY = "keywords" ORIGIN_METADATA_VALUE = "git" ORIGIN_MASTER_REVISION = {} ORIGIN_MASTER_DIRECTORY = {} def _add_origin( storage, search, counters, origin_url, visit_type="git", snapshot_branches={} ): storage.origin_add([Origin(url=origin_url)]) search.origin_update( [{"url": origin_url, "has_visits": True, "visit_types": [visit_type]}] ) counters.add("origin", [origin_url]) date = now() visit = OriginVisit(origin=origin_url, date=date, type=visit_type) visit = storage.origin_visit_add([visit])[0] counters.add("origin_visit", [f"{visit.unique_key()}"]) snapshot = Snapshot.from_dict({"branches": snapshot_branches}) storage.snapshot_add([snapshot]) counters.add("snapshot", [snapshot.id]) visit_status = OriginVisitStatus( origin=origin_url, visit=visit.visit, date=date + timedelta(minutes=1), type=visit.type, status="full", snapshot=snapshot.id, ) storage.origin_visit_status_add([visit_status]) counters.add("origin_visit_status", [f"{visit_status.unique_key()}"]) # Tests data initialization def _init_tests_data(): # To hold reference to the memory storage storage = get_storage("memory") # Create search instance search = get_search("memory") search.initialize() search.origin_update({"url": origin["url"]} for origin in _TEST_ORIGINS) # create the counters instance counters = get_counters("memory") # Create indexer storage instance that will be shared by indexers idx_storage = get_indexer_storage("memory") # Declare a test tool for origin intrinsic metadata tests idx_tool = idx_storage.indexer_configuration_add([INDEXER_TOOL])[0] INDEXER_TOOL["id"] = idx_tool["id"] # Load git repositories from archives for origin in _TEST_ORIGINS: for i, archive_ in enumerate(origin["archives"]): if i > 0: # ensure visit dates will be different when simulating # multiple visits of an origin time.sleep(1) origin_repo_archive = os.path.join( os.path.dirname(__file__), "resources/repos/%s" % archive_ ) loader = GitLoaderFromArchive( storage, origin["url"], archive_path=origin_repo_archive, ) result = loader.load() assert result["status"] == "eventful" ori = storage.origin_get([origin["url"]])[0] origin.update(ori.to_dict()) # add an 'id' key if enabled search.origin_update( [{"url": origin["url"], "has_visits": True, "visit_types": ["git"]}] ) for i in range(250): _add_origin( storage, search, counters, origin_url=f"https://many.origins/{i+1}", visit_type="tar", ) sha1s: Set[Sha1] = set() directories = set() revisions = set() releases = set() snapshots = set() swhids = [] content_path = {} # Get all objects loaded into the test archive common_metadata = {ORIGIN_METADATA_KEY: ORIGIN_METADATA_VALUE} for origin in _TEST_ORIGINS: origin_revisions = set() snp = snapshot_get_latest(storage, origin["url"]) swhids.append( QualifiedSWHID( object_type=ObjectType.SNAPSHOT, object_id=snp.id, origin=origin["url"] ) ) snapshots.add(hash_to_hex(snp.id)) for branch_name, branch_data in snp.branches.items(): target_type = branch_data.target_type.value if target_type == "revision": origin_revisions.add(branch_data.target) swhids.append( QualifiedSWHID( object_type=ObjectType.REVISION, object_id=branch_data.target, origin=origin["url"], visit=CoreSWHID( object_type=ObjectType.SNAPSHOT, object_id=snp.id ), ) ) if b"master" in branch_name: # Add some origin intrinsic metadata for tests metadata = common_metadata metadata.update(origin.get("metadata", {})) revision = storage.revision_get([branch_data.target])[0] origin_metadata = OriginIntrinsicMetadataRow( id=origin["url"], from_directory=revision.directory, indexer_configuration_id=idx_tool["id"], metadata=metadata, mappings=[], ) idx_storage.origin_intrinsic_metadata_add([origin_metadata]) search.origin_update([{"url": origin["url"], "jsonld": metadata}]) ORIGIN_MASTER_REVISION[origin["url"]] = hash_to_hex( branch_data.target ) ORIGIN_MASTER_DIRECTORY[origin["url"]] = hash_to_hex( revision.directory ) elif target_type == "release": release = storage.release_get([branch_data.target])[0] origin_revisions.add(release.target) releases.add(hash_to_hex(branch_data.target)) swhids.append( QualifiedSWHID( object_type=ObjectType.RELEASE, object_id=branch_data.target, origin=origin["url"], visit=CoreSWHID( object_type=ObjectType.SNAPSHOT, object_id=snp.id ), ) ) for rev_log in storage.revision_shortlog(origin_revisions): rev_id = rev_log[0] revisions.add(rev_id) for rev in storage.revision_get(sorted(origin_revisions)): if rev is None: continue dir_id = rev.directory directories.add(hash_to_hex(dir_id)) for entry in dir_iterator(storage, dir_id): if entry["type"] == "file": sha1s.add(entry["sha1"]) content_path[entry["sha1"]] = "/".join( [hash_to_hex(dir_id), entry["path"].decode("utf-8")] ) swhids.append( QualifiedSWHID( object_type=ObjectType.CONTENT, object_id=entry["sha1_git"], origin=origin["url"], visit=CoreSWHID( object_type=ObjectType.SNAPSHOT, object_id=snp.id ), anchor=CoreSWHID( object_type=ObjectType.REVISION, object_id=rev.id ), path=b"/" + entry["path"], ) ) elif entry["type"] == "dir": directories.add(hash_to_hex(entry["target"])) swhids.append( QualifiedSWHID( object_type=ObjectType.DIRECTORY, object_id=entry["target"], origin=origin["url"], visit=CoreSWHID( object_type=ObjectType.SNAPSHOT, object_id=snp.id ), anchor=CoreSWHID( object_type=ObjectType.REVISION, object_id=rev.id ), path=b"/" + entry["path"] + b"/", ) ) _add_extra_contents(storage, sha1s) # Get all checksums for each content result: List[Optional[Content]] = storage.content_get(list(sha1s)) contents: List[Dict] = [] for content in result: assert content is not None sha1 = hash_to_hex(content.sha1) content_metadata = { algo: hash_to_hex(getattr(content, algo)) for algo in DEFAULT_ALGORITHMS } path = "" if content.sha1 in content_path: path = content_path[content.sha1] cnt_data = storage.content_get_data(content.sha1) assert cnt_data is not None mimetype, encoding = get_mimetype_and_encoding_for_content(cnt_data) _, _, cnt_data = re_encode_content(mimetype, encoding, cnt_data) content_display_data = prepare_content_for_display(cnt_data, mimetype, path) content_metadata.update( { "path": path, "mimetype": mimetype, "encoding": encoding, "hljs_language": content_display_data["language"], "raw_data": cnt_data, "data": content_display_data["content_data"], } ) _contents[sha1] = content_metadata contents.append(content_metadata) # Add the empty directory to the test archive storage.directory_add([Directory(entries=())]) # Add empty content to the test archive storage.content_add([Content.from_data(data=b"")]) # Add fake git origin with pull request branches _add_origin( storage, search, counters, origin_url="https://git.example.org/project", snapshot_branches={ b"refs/heads/master": { "target_type": "revision", "target": next(iter(revisions)), }, **{ f"refs/pull/{i}".encode(): { "target_type": "revision", "target": next(iter(revisions)), } for i in range(300) }, }, ) counters.add("revision", revisions) counters.add("release", releases) counters.add("directory", directories) counters.add("content", [content["sha1"] for content in contents]) # Return tests data return { "search": search, "storage": storage, "idx_storage": idx_storage, "counters": counters, "origins": _TEST_ORIGINS, "contents": list(sorted(contents, key=lambda c: c["sha1"])), "directories": list(sorted(directories)), "releases": list(sorted(releases)), "revisions": list(sorted(map(hash_to_hex, revisions))), "snapshots": list(sorted(snapshots)), "swhids": swhids, } def _init_indexers(tests_data): # Instantiate content indexers that will be used in tests # and force them to use the memory storages indexers = {} for idx_name, idx_class, idx_config in ( ("mimetype_indexer", MimetypeIndexer, _TEST_MIMETYPE_INDEXER_CONFIG), ("license_indexer", FossologyLicenseIndexer, _TEST_LICENSE_INDEXER_CONFIG), - ("ctags_indexer", CtagsIndexer, _TEST_CTAGS_INDEXER_CONFIG), ): idx = idx_class(config=idx_config) idx.storage = tests_data["storage"] idx.objstorage = tests_data["storage"].objstorage idx.idx_storage = tests_data["idx_storage"] idx.register_tools(idx.config["tools"]) indexers[idx_name] = idx return indexers def get_content(content_sha1): return _contents.get(content_sha1) _tests_data = None _current_tests_data = None _indexer_loggers = {} def get_tests_data(reset=False): """ Initialize tests data and return them in a dict. """ global _tests_data, _current_tests_data if _tests_data is None: _tests_data = _init_tests_data() indexers = _init_indexers(_tests_data) for (name, idx) in indexers.items(): # pytest makes the loggers use a temporary file; and deepcopy # requires serializability. So we remove them, and add them # back after the copy. _indexer_loggers[name] = idx.log del idx.log _tests_data.update(indexers) if reset or _current_tests_data is None: _current_tests_data = deepcopy(_tests_data) for (name, logger) in _indexer_loggers.items(): _current_tests_data[name].log = logger return _current_tests_data def override_storages(storage, idx_storage, search, counters): """ Helper function to replace the storages from which archive data are fetched. """ swh_config = config.get_config() swh_config.update( { "storage": storage, "indexer_storage": idx_storage, "search": search, "counters": counters, } ) archive.storage = storage archive.idx_storage = idx_storage archive.search = search archive.counters = counters