diff --git a/swh/web/common/archive.py b/swh/web/common/archive.py index 3f9824c5..2547c8da 100644 --- a/swh/web/common/archive.py +++ b/swh/web/common/archive.py @@ -1,1421 +1,1385 @@ # Copyright (C) 2015-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from collections import defaultdict import itertools import os import re from typing import Any, Dict, Iterable, Iterator, List, Optional, Set, Tuple, Union from urllib.parse import urlparse from swh.model import hashutil from swh.model.identifiers import CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT from swh.model.model import OriginVisit, Revision from swh.storage.algos import diff, revisions_walker from swh.storage.algos.origin import origin_get_latest_visit_status from swh.storage.algos.snapshot import snapshot_get_latest, snapshot_resolve_alias from swh.vault.exc import NotFoundExc as VaultNotFoundExc from swh.web import config from swh.web.common import converters, query from swh.web.common.exc import BadInputExc, NotFoundExc from swh.web.common.typing import ( OriginInfo, OriginMetadataInfo, OriginVisitInfo, PagedResult, ) search = config.search() storage = config.storage() vault = config.vault() idx_storage = config.indexer_storage() MAX_LIMIT = 50 # Top limit the users can ask for def _first_element(lst): """Returns the first element in the provided list or None if it is empty or None""" return next(iter(lst or []), None) def lookup_multiple_hashes(hashes): """Lookup the passed hashes in a single DB connection, using batch processing. Args: An array of {filename: X, sha1: Y}, string X, hex sha1 string Y. Returns: The same array with elements updated with elem['found'] = true if the hash is present in storage, elem['found'] = false if not. """ hashlist = [hashutil.hash_to_bytes(elem["sha1"]) for elem in hashes] content_missing = storage.content_missing_per_sha1(hashlist) missing = [hashutil.hash_to_hex(x) for x in content_missing] for x in hashes: x.update({"found": True}) for h in hashes: if h["sha1"] in missing: h["found"] = False return hashes def lookup_expression(expression, last_sha1, per_page): """Lookup expression in raw content. Args: expression (str): An expression to lookup through raw indexed content last_sha1 (str): Last sha1 seen per_page (int): Number of results per page Yields: ctags whose content match the expression """ limit = min(per_page, MAX_LIMIT) ctags = idx_storage.content_ctags_search( expression, last_sha1=last_sha1, limit=limit ) for ctag in ctags: ctag = converters.from_swh(ctag, hashess={"id"}) ctag["sha1"] = ctag["id"] ctag.pop("id") yield ctag def lookup_hash(q: str) -> Dict[str, Any]: """Check if the storage contains a given content checksum and return it if found. Args: q: query string of the form Returns: Dict with key found containing the hash info if the hash is present, None if not. """ algo, hash_ = query.parse_hash(q) found = _first_element(storage.content_find({algo: hash_})) if found: content = converters.from_content(found.to_dict()) else: content = None return {"found": content, "algo": algo} def search_hash(q: str) -> Dict[str, bool]: """Search storage for a given content checksum. Args: q: query string of the form Returns: Dict with key found to True or False, according to whether the checksum is present or not """ algo, hash_ = query.parse_hash(q) found = _first_element(storage.content_find({algo: hash_})) return {"found": found is not None} def _lookup_content_sha1(q: str) -> Optional[bytes]: """Given a possible input, query for the content's sha1. Args: q: query string of the form Returns: binary sha1 if found or None """ algo, hash_ = query.parse_hash(q) if algo != "sha1": hashes = _first_element(storage.content_find({algo: hash_})) if not hashes: return None return hashes.sha1 return hash_ def lookup_content_ctags(q): """Return ctags information from a specified content. Args: q: query string of the form Yields: ctags information (dict) list if the content is found. """ sha1 = _lookup_content_sha1(q) if not sha1: return None ctags = list(idx_storage.content_ctags_get([sha1])) if not ctags: return None for ctag in ctags: yield converters.from_swh(ctag, hashess={"id"}) def lookup_content_filetype(q): """Return filetype information from a specified content. Args: q: query string of the form Yields: filetype information (dict) list if the content is found. """ sha1 = _lookup_content_sha1(q) if not sha1: return None filetype = _first_element(list(idx_storage.content_mimetype_get([sha1]))) if not filetype: return None return converters.from_filetype(filetype.to_dict()) def lookup_content_language(q): """Always returns None. This used to return language information from a specified content, but this is currently disabled. Args: q: query string of the form Yields: language information (dict) list if the content is found. """ return None def lookup_content_license(q): """Return license information from a specified content. Args: q: query string of the form Yields: license information (dict) list if the content is found. """ sha1 = _lookup_content_sha1(q) if not sha1: return None licenses = list(idx_storage.content_fossology_license_get([sha1])) if not licenses: return None license_dicts = [license.to_dict() for license in licenses] for license_dict in license_dicts: del license_dict["id"] lic = { "id": sha1, "facts": license_dicts, } return converters.from_swh(lic, hashess={"id"}) def lookup_origin(origin: OriginInfo) -> OriginInfo: """Return information about the origin matching dict origin. Args: origin: origin's dict with 'url' key Returns: origin information as dict. """ origin_urls = [origin["url"]] if origin["url"]: # handle case when user provided an origin url with a trailing # slash while the url in storage does not have it (e.g. GitHub) if origin["url"].endswith("/"): origin_urls.append(origin["url"][:-1]) # handle case when user provided an origin url without a trailing # slash while the url in storage have it (e.g. Debian source package) else: origin_urls.append(f"{origin['url']}/") try: # handle case where the "://" character sequence was mangled into ":/" parsed_url = urlparse(origin["url"]) if ( parsed_url.scheme and not parsed_url.netloc and origin["url"].startswith(f"{parsed_url.scheme}:/") and not origin["url"].startswith(f"{parsed_url.scheme}://") ): origin_urls.append( origin["url"].replace( f"{parsed_url.scheme}:/", f"{parsed_url.scheme}://" ) ) except Exception: pass origins = [o for o in storage.origin_get(origin_urls) if o is not None] if not origins: msg = "Origin with url %s not found!" % origin["url"] raise NotFoundExc(msg) return converters.from_origin(origins[0].to_dict()) def lookup_origins( page_token: Optional[str], limit: int = 100 ) -> PagedResult[OriginInfo]: """Get list of archived software origins in a paginated way. Origins are sorted by id before returning them Args: origin_from (int): The minimum id of the origins to return origin_count (int): The maximum number of origins to return Returns: Page of OriginInfo """ page = storage.origin_list(page_token=page_token, limit=limit) return PagedResult( [converters.from_origin(o.to_dict()) for o in page.results], next_page_token=page.next_page_token, ) def search_origin( url_pattern: str, 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 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: 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()["metadata_search_backend"] == "swh-search": page_result = search.origin_search(metadata_pattern=fulltext, limit=limit,) matches = idx_storage.origin_intrinsic_metadata_get( [r["url"] for r in page_result.results] ) else: matches = idx_storage.origin_intrinsic_metadata_search_fulltext( conjunction=[fulltext], limit=limit ) matches = [match.to_dict() for match in matches] origins = storage.origin_get([match["id"] for match in matches]) for origin, match in zip(origins, matches): if not origin: continue match["from_revision"] = hashutil.hash_to_hex(match["from_revision"]) del match["id"] results.append(OriginMetadataInfo(url=origin.url, metadata=match)) return results def lookup_origin_intrinsic_metadata(origin_url: str) -> Dict[str, Any]: """Return intrinsic metadata for origin whose origin matches given origin. Args: origin_url: origin url Raises: NotFoundExc when the origin is not found Returns: origin metadata. """ origins = [origin_url] origin_info = storage.origin_get(origins)[0] if not origin_info: raise NotFoundExc(f"Origin with url {origin_url} not found!") match = _first_element(idx_storage.origin_intrinsic_metadata_get(origins)) result = {} if match: result = match.metadata return result def _to_sha1_bin(sha1_hex): _, sha1_git_bin = query.parse_hash_with_algorithms_or_throws( sha1_hex, ["sha1"], "Only sha1_git is supported." # HACK: sha1_git really ) return sha1_git_bin def _check_directory_exists(sha1_git, sha1_git_bin): if len(list(storage.directory_missing([sha1_git_bin]))): raise NotFoundExc("Directory with sha1_git %s not found" % sha1_git) def lookup_directory(sha1_git): """Return information about the directory with id sha1_git. Args: sha1_git as string Returns: directory information as dict. """ empty_dir_sha1 = "4b825dc642cb6eb9a060e54bf8d69288fbee4904" if sha1_git == empty_dir_sha1: return [] sha1_git_bin = _to_sha1_bin(sha1_git) _check_directory_exists(sha1_git, sha1_git_bin) directory_entries = storage.directory_ls(sha1_git_bin) return map(converters.from_directory_entry, directory_entries) def lookup_directory_with_path(sha1_git: str, path: str) -> Dict[str, Any]: """Return directory information for entry with specified path w.r.t. root directory pointed by sha1_git Args: sha1_git: sha1_git corresponding to the directory to which we append paths to (hopefully) find the entry path: the relative path to the entry starting from the root directory pointed by sha1_git Returns: Directory entry information as dict. Raises: NotFoundExc if the directory entry is not found """ sha1_git_bin = _to_sha1_bin(sha1_git) _check_directory_exists(sha1_git, sha1_git_bin) paths = path.strip(os.path.sep).split(os.path.sep) queried_dir = storage.directory_entry_get_by_path( sha1_git_bin, [p.encode("utf-8") for p in paths] ) if not queried_dir: raise NotFoundExc( f"Directory entry with path {path} from root directory {sha1_git} not found" ) return converters.from_directory_entry(queried_dir) def lookup_release(release_sha1_git: str) -> Dict[str, Any]: """Return information about the release with sha1 release_sha1_git. Args: release_sha1_git: The release's sha1 as hexadecimal Returns: Release information as dict. Raises: ValueError if the identifier provided is not of sha1 nature. """ sha1_git_bin = _to_sha1_bin(release_sha1_git) release = _first_element(storage.release_get([sha1_git_bin])) if not release: raise NotFoundExc(f"Release with sha1_git {release_sha1_git} not found.") return converters.from_release(release) def lookup_release_multiple(sha1_git_list) -> Iterator[Optional[Dict[str, Any]]]: """Return information about the releases identified with their sha1_git identifiers. Args: sha1_git_list: A list of release sha1_git identifiers Returns: Iterator of Release metadata information as dict. Raises: ValueError if the identifier provided is not of sha1 nature. """ sha1_bin_list = [_to_sha1_bin(sha1_git) for sha1_git in sha1_git_list] releases = storage.release_get(sha1_bin_list) for r in releases: if r is not None: yield converters.from_release(r) else: yield None def lookup_revision(rev_sha1_git) -> Dict[str, Any]: """Return information about the revision with sha1 revision_sha1_git. Args: revision_sha1_git: The revision's sha1 as hexadecimal Returns: Revision information as dict. Raises: ValueError if the identifier provided is not of sha1 nature. NotFoundExc if there is no revision with the provided sha1_git. """ sha1_git_bin = _to_sha1_bin(rev_sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision with sha1_git {rev_sha1_git} not found.") return converters.from_revision(revision) def lookup_revision_multiple(sha1_git_list) -> Iterator[Optional[Dict[str, Any]]]: """Return information about the revisions identified with their sha1_git identifiers. Args: sha1_git_list: A list of revision sha1_git identifiers Yields: revision information as dict if the revision exists, None otherwise. Raises: ValueError if the identifier provided is not of sha1 nature. """ sha1_bin_list = [_to_sha1_bin(sha1_git) for sha1_git in sha1_git_list] revisions = storage.revision_get(sha1_bin_list) for revision in revisions: if revision is not None: yield converters.from_revision(revision) else: yield None def lookup_revision_message(rev_sha1_git) -> Dict[str, bytes]: """Return the raw message of the revision with sha1 revision_sha1_git. Args: revision_sha1_git: The revision's sha1 as hexadecimal Returns: Decoded revision message as dict {'message': } Raises: ValueError if the identifier provided is not of sha1 nature. NotFoundExc if the revision is not found, or if it has no message """ sha1_git_bin = _to_sha1_bin(rev_sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision with sha1_git {rev_sha1_git} not found.") if not revision.message: raise NotFoundExc(f"No message for revision with sha1_git {rev_sha1_git}.") return {"message": revision.message} def _lookup_revision_id_by(origin, branch_name, timestamp): def _get_snapshot_branch(snapshot, branch_name): snapshot = lookup_snapshot( visit["snapshot"], branches_from=branch_name, branches_count=10, branch_name_exclude_prefix=None, ) branch = None if branch_name in snapshot["branches"]: branch = snapshot["branches"][branch_name] return branch if isinstance(origin, int): origin = {"id": origin} elif isinstance(origin, str): origin = {"url": origin} else: raise TypeError('"origin" must be an int or a string.') from swh.web.common.origin_visits import get_origin_visit visit = get_origin_visit(origin, visit_ts=timestamp) branch = _get_snapshot_branch(visit["snapshot"], branch_name) rev_id = None if branch and branch["target_type"] == "revision": rev_id = branch["target"] elif branch and branch["target_type"] == "alias": branch = _get_snapshot_branch(visit["snapshot"], branch["target"]) if branch and branch["target_type"] == "revision": rev_id = branch["target"] if not rev_id: raise NotFoundExc( "Revision for origin %s and branch %s not found." % (origin.get("url"), branch_name) ) return rev_id def lookup_revision_by(origin, branch_name="HEAD", timestamp=None): """Lookup revision by origin, snapshot branch name and visit timestamp. If branch_name is not provided, lookup using 'HEAD' as default. If timestamp is not provided, use the most recent. Args: origin (Union[int,str]): origin of the revision branch_name (str): snapshot branch name timestamp (str/int): origin visit time frame Returns: dict: The revision matching the criterions Raises: NotFoundExc if no revision corresponds to the criterion """ rev_id = _lookup_revision_id_by(origin, branch_name, timestamp) return lookup_revision(rev_id) def lookup_revision_log(rev_sha1_git, limit): """Lookup revision log by revision id. Args: rev_sha1_git (str): The revision's sha1 as hexadecimal limit (int): the maximum number of revisions returned Returns: list: Revision log as list of revision dicts Raises: ValueError: if the identifier provided is not of sha1 nature. swh.web.common.exc.NotFoundExc: if there is no revision with the provided sha1_git. """ lookup_revision(rev_sha1_git) sha1_git_bin = _to_sha1_bin(rev_sha1_git) revision_entries = storage.revision_log([sha1_git_bin], limit) return map(converters.from_revision, revision_entries) def lookup_revision_log_by(origin, branch_name, timestamp, limit): """Lookup revision by origin, snapshot branch name and visit timestamp. Args: origin (Union[int,str]): origin of the revision branch_name (str): snapshot branch timestamp (str/int): origin visit time frame limit (int): the maximum number of revisions returned Returns: list: Revision log as list of revision dicts Raises: swh.web.common.exc.NotFoundExc: if no revision corresponds to the criterion """ rev_id = _lookup_revision_id_by(origin, branch_name, timestamp) return lookup_revision_log(rev_id, limit) def lookup_revision_with_context_by( origin, branch_name, timestamp, sha1_git, limit=100 ): """Return information about revision sha1_git, limited to the sub-graph of all transitive parents of sha1_git_root. sha1_git_root being resolved through the lookup of a revision by origin, branch_name and ts. In other words, sha1_git is an ancestor of sha1_git_root. Args: - origin: origin of the revision. - branch_name: revision's branch. - timestamp: revision's time frame. - sha1_git: one of sha1_git_root's ancestors. - limit: limit the lookup to 100 revisions back. Returns: Pair of (root_revision, revision). Information on sha1_git if it is an ancestor of sha1_git_root including children leading to sha1_git_root Raises: - BadInputExc in case of unknown algo_hash or bad hash. - NotFoundExc if either revision is not found or if sha1_git is not an ancestor of sha1_git_root. """ rev_root_id = _lookup_revision_id_by(origin, branch_name, timestamp) rev_root_id_bin = hashutil.hash_to_bytes(rev_root_id) rev_root = storage.revision_get([rev_root_id_bin])[0] return ( converters.from_revision(rev_root) if rev_root else None, lookup_revision_with_context(rev_root, sha1_git, limit), ) def lookup_revision_with_context( sha1_git_root: Union[str, Dict[str, Any], Revision], sha1_git: str, limit: int = 100 ) -> Dict[str, Any]: """Return information about revision sha1_git, limited to the sub-graph of all transitive parents of sha1_git_root. In other words, sha1_git is an ancestor of sha1_git_root. Args: sha1_git_root: latest revision. The type is either a sha1 (as an hex string) or a non converted dict. sha1_git: one of sha1_git_root's ancestors limit: limit the lookup to 100 revisions back Returns: Information on sha1_git if it is an ancestor of sha1_git_root including children leading to sha1_git_root Raises: BadInputExc in case of unknown algo_hash or bad hash NotFoundExc if either revision is not found or if sha1_git is not an ancestor of sha1_git_root """ sha1_git_bin = _to_sha1_bin(sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision {sha1_git} not found") if isinstance(sha1_git_root, str): sha1_git_root_bin = _to_sha1_bin(sha1_git_root) revision_root = storage.revision_get([sha1_git_root_bin])[0] if not revision_root: raise NotFoundExc(f"Revision root {sha1_git_root} not found") elif isinstance(sha1_git_root, Revision): sha1_git_root_bin = sha1_git_root.id else: sha1_git_root_bin = sha1_git_root["id"] revision_log = storage.revision_log([sha1_git_root_bin], limit) parents: Dict[str, List[str]] = {} children = defaultdict(list) for rev in revision_log: rev_id = rev["id"] parents[rev_id] = [] for parent_id in rev["parents"]: parents[rev_id].append(parent_id) children[parent_id].append(rev_id) if revision.id not in parents: raise NotFoundExc(f"Revision {sha1_git} is not an ancestor of {sha1_git_root}") revision_d = revision.to_dict() revision_d["children"] = children[revision.id] return converters.from_revision(revision_d) def lookup_directory_with_revision(sha1_git, dir_path=None, with_data=False): """Return information on directory pointed by revision with sha1_git. If dir_path is not provided, display top level directory. Otherwise, display the directory pointed by dir_path (if it exists). Args: sha1_git: revision's hash. dir_path: optional directory pointed to by that revision. with_data: boolean that indicates to retrieve the raw data if the path resolves to a content. Default to False (for the api) Returns: Information on the directory pointed to by that revision. Raises: BadInputExc in case of unknown algo_hash or bad hash. NotFoundExc either if the revision is not found or the path referenced does not exist. NotImplementedError in case of dir_path exists but do not reference a type 'dir' or 'file'. """ sha1_git_bin = _to_sha1_bin(sha1_git) revision = storage.revision_get([sha1_git_bin])[0] if not revision: raise NotFoundExc(f"Revision {sha1_git} not found") dir_sha1_git_bin = revision.directory if dir_path: paths = dir_path.strip(os.path.sep).split(os.path.sep) entity = storage.directory_entry_get_by_path( dir_sha1_git_bin, list(map(lambda p: p.encode("utf-8"), paths)) ) if not entity: raise NotFoundExc( "Directory or File '%s' pointed to by revision %s not found" % (dir_path, sha1_git) ) else: entity = {"type": "dir", "target": dir_sha1_git_bin} if entity["type"] == "dir": directory_entries = storage.directory_ls(entity["target"]) or [] return { "type": "dir", "path": "." if not dir_path else dir_path, "revision": sha1_git, "content": list(map(converters.from_directory_entry, directory_entries)), } elif entity["type"] == "file": # content content = _first_element(storage.content_find({"sha1_git": entity["target"]})) if not content: raise NotFoundExc(f"Content not found for revision {sha1_git}") content_d = content.to_dict() if with_data: data = storage.content_get_data(content.sha1) if data: content_d["data"] = data return { "type": "file", "path": "." if not dir_path else dir_path, "revision": sha1_git, "content": converters.from_content(content_d), } elif entity["type"] == "rev": # revision revision = storage.revision_get([entity["target"]])[0] return { "type": "rev", "path": "." if not dir_path else dir_path, "revision": sha1_git, "content": converters.from_revision(revision) if revision else None, } else: raise NotImplementedError("Entity of type %s not implemented." % entity["type"]) def lookup_content(q: str) -> Dict[str, Any]: """Lookup the content designed by q. Args: q: The release's sha1 as hexadecimal Raises: NotFoundExc if the requested content is not found """ algo, hash_ = query.parse_hash(q) c = _first_element(storage.content_find({algo: hash_})) if not c: hhex = hashutil.hash_to_hex(hash_) raise NotFoundExc(f"Content with {algo} checksum equals to {hhex} not found!") return converters.from_content(c.to_dict()) def lookup_content_raw(q: str) -> Dict[str, Any]: """Lookup the content defined by q. Args: q: query string of the form Returns: dict with 'sha1' and 'data' keys. data representing its raw data decoded. Raises: NotFoundExc if the requested content is not found or if the content bytes are not available in the storage """ c = lookup_content(q) content_sha1_bytes = hashutil.hash_to_bytes(c["checksums"]["sha1"]) content_data = storage.content_get_data(content_sha1_bytes) if content_data is None: algo, hash_ = query.parse_hash(q) raise NotFoundExc( f"Bytes of content with {algo} checksum equals " f"to {hashutil.hash_to_hex(hash_)} are not available!" ) return converters.from_content({"sha1": content_sha1_bytes, "data": content_data}) def stat_counters(): """Return the stat counters for Software Heritage Returns: A dict mapping textual labels to integer values. """ return storage.stat_counters() def _lookup_origin_visits( origin_url: str, last_visit: Optional[int] = None, limit: int = 10 ) -> Iterator[OriginVisit]: """Yields the origin origins' visits. Args: origin_url (str): origin to list visits for last_visit (int): last visit to lookup from limit (int): Number of elements max to display Yields: OriginVisit for that origin """ limit = min(limit, MAX_LIMIT) page_token: Optional[str] if last_visit is not None: page_token = str(last_visit) else: page_token = None visit_page = storage.origin_visit_get( origin_url, page_token=page_token, limit=limit ) yield from visit_page.results def lookup_origin_visits( origin: str, last_visit: Optional[int] = None, per_page: int = 10 ) -> Iterator[OriginVisitInfo]: """Yields the origin origins' visits. Args: origin: origin to list visits for Yields: Dictionaries of origin_visit for that origin """ for visit in _lookup_origin_visits(origin, last_visit=last_visit, limit=per_page): visit_status = storage.origin_visit_status_get_latest(origin, visit.visit) yield converters.from_origin_visit( {**visit_status.to_dict(), "type": visit.type} ) def lookup_origin_visit_latest( origin_url: str, require_snapshot: bool = False, type: Optional[str] = None, allowed_statuses: Optional[List[str]] = None, ) -> Optional[OriginVisitInfo]: """Return the origin's latest visit Args: origin_url: origin to list visits for type: Optional visit type to filter on (e.g git, tar, dsc, svn, hg, npm, pypi, ...) allowed_statuses: list of visit statuses considered to find the latest visit. For instance, ``allowed_statuses=['full']`` will only consider visits that have successfully run to completion. require_snapshot: filter out origins without a snapshot Returns: The origin visit info as dict if found """ visit_status = origin_get_latest_visit_status( storage, origin_url, type=type, allowed_statuses=allowed_statuses, require_snapshot=require_snapshot, ) return ( converters.from_origin_visit(visit_status.to_dict()) if visit_status else None ) def lookup_origin_visit(origin_url: str, visit_id: int) -> OriginVisitInfo: """Return information about visit visit_id with origin origin. Args: origin: origin concerned by the visit visit_id: the visit identifier to lookup Yields: The dict origin_visit concerned """ visit = storage.origin_visit_get_by(origin_url, visit_id) visit_status = storage.origin_visit_status_get_latest(origin_url, visit_id) if not visit: raise NotFoundExc( f"Origin {origin_url} or its visit with id {visit_id} not found!" ) return converters.from_origin_visit({**visit_status.to_dict(), "type": visit.type}) def lookup_snapshot_sizes( snapshot_id: str, branch_name_exclude_prefix: Optional[str] = "refs/pull/" ) -> Dict[str, int]: """Count the number of branches in the snapshot with the given id Args: snapshot_id (str): sha1 identifier of the snapshot Returns: dict: A dict whose keys are the target types of branches and values their corresponding amount """ snapshot_id_bin = _to_sha1_bin(snapshot_id) snapshot_sizes = dict.fromkeys(("alias", "release", "revision"), 0) branch_counts = storage.snapshot_count_branches( snapshot_id_bin, branch_name_exclude_prefix.encode() if branch_name_exclude_prefix else None, ) # remove possible None key returned by snapshot_count_branches # when null branches are present in the snapshot branch_counts.pop(None, None) snapshot_sizes.update(branch_counts) return snapshot_sizes def lookup_snapshot( snapshot_id: str, branches_from: str = "", branches_count: int = 1000, target_types: Optional[List[str]] = None, branch_name_include_substring: Optional[str] = None, branch_name_exclude_prefix: Optional[str] = "refs/pull/", ) -> Dict[str, Any]: """Return information about a snapshot, aka the list of named branches found during a specific visit of an origin. Args: snapshot_id: sha1 identifier of the snapshot branches_from: optional parameter used to skip branches whose name is lesser than it before returning them branches_count: optional parameter used to restrain the amount of returned branches target_types: optional parameter used to filter the target types of branch to return (possible values that can be contained in that list are `'content', 'directory', 'revision', 'release', 'snapshot', 'alias'`) branch_name_include_substring: if provided, only return branches whose name contains given substring branch_name_exclude_prefix: if provided, do not return branches whose name starts with given pattern Returns: A dict filled with the snapshot content. """ snapshot_id_bin = _to_sha1_bin(snapshot_id) 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, ) if not partial_branches: raise NotFoundExc(f"Snapshot with id {snapshot_id} not found!") return converters.from_partial_branches(partial_branches) 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_branch_name_from_tip_revision( - snapshot_id: str, revision_id: str -) -> Optional[str]: - """Check if a revision corresponds to the tip of a snapshot branch - - Args: - snapshot_id: hexadecimal representation of a snapshot id - revision_id: hexadecimal representation of a revision id - - Returns: - The name of the first found branch or None otherwise - """ - per_page = 10000 - branches_from = "" - snapshot: Dict[str, Any] = {"branches": {}} - branches = [] - while not branches_from or len(snapshot["branches"]) == per_page + 1: - snapshot = lookup_snapshot( - snapshot_id, - target_types=[REVISION], - branches_from=branches_from, - branches_count=per_page + 1, - branch_name_exclude_prefix=None, - ) - - branches += [ - {"name": k, "target": v["target"]} for k, v in snapshot["branches"].items() - ] - branches_from = branches[-1]["name"] - - for branch in branches: - if branch["target"] == revision_id: - return branch["name"] - return 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, *args, **kwargs): try: return vault_fn(*args, **kwargs) except VaultNotFoundExc: return None def vault_cook(obj_type, obj_id, email=None): """Cook a vault bundle. """ return _vault_request(vault.cook, obj_type, obj_id, email=email) def vault_fetch(obj_type, obj_id): """Fetch a vault bundle. """ return _vault_request(vault.fetch, obj_type, obj_id) def vault_progress(obj_type, obj_id): """Get the current progress of a vault bundle. """ return _vault_request(vault.progress, obj_type, obj_id) def diff_revision(rev_id): """Get the list of file changes (insertion / deletion / modification / renaming) for a particular revision. """ rev_sha1_git_bin = _to_sha1_bin(rev_id) changes = diff.diff_revision(storage, rev_sha1_git_bin, track_renaming=True) for change in changes: change["from"] = converters.from_directory_entry(change["from"]) change["to"] = converters.from_directory_entry(change["to"]) if change["from_path"]: change["from_path"] = change["from_path"].decode("utf-8") if change["to_path"]: change["to_path"] = change["to_path"].decode("utf-8") return changes class _RevisionsWalkerProxy(object): """ Proxy class wrapping a revisions walker iterator from swh-storage and performing needed conversions. """ def __init__(self, rev_walker_type, rev_start, *args, **kwargs): rev_start_bin = hashutil.hash_to_bytes(rev_start) self.revisions_walker = revisions_walker.get_revisions_walker( rev_walker_type, storage, rev_start_bin, *args, **kwargs ) def export_state(self): return self.revisions_walker.export_state() def __next__(self): return converters.from_revision(next(self.revisions_walker)) def __iter__(self): return self def get_revisions_walker(rev_walker_type, rev_start, *args, **kwargs): """ Utility function to instantiate a revisions walker of a given type, see :mod:`swh.storage.algos.revisions_walker`. Args: rev_walker_type (str): the type of revisions walker to return, possible values are: ``committer_date``, ``dfs``, ``dfs_post``, ``bfs`` and ``path`` rev_start (str): hexadecimal representation of a revision identifier args (list): position arguments to pass to the revisions walker constructor kwargs (dict): keyword arguments to pass to the revisions walker constructor """ # first check if the provided revision is valid lookup_revision(rev_start) return _RevisionsWalkerProxy(rev_walker_type, rev_start, *args, **kwargs) def lookup_object(object_type: str, object_id: str) -> Dict[str, Any]: """ Utility function for looking up an object in the archive by its type and id. Args: object_type (str): the type of object to lookup, either *content*, *directory*, *release*, *revision* or *snapshot* object_id (str): the *sha1_git* checksum identifier in hexadecimal form of the object to lookup Returns: Dict[str, Any]: A dictionary describing the object or a list of dictionary for the directory object type. Raises: swh.web.common.exc.NotFoundExc: if the object could not be found in the archive BadInputExc: if the object identifier is invalid """ if object_type == CONTENT: return lookup_content(f"sha1_git:{object_id}") elif object_type == DIRECTORY: return {"id": object_id, "content": list(lookup_directory(object_id))} elif object_type == RELEASE: return lookup_release(object_id) elif object_type == REVISION: return lookup_revision(object_id) elif object_type == SNAPSHOT: return lookup_snapshot(object_id) raise BadInputExc( ( "Invalid swh object type! Valid types are " f"{CONTENT}, {DIRECTORY}, {RELEASE} " f"{REVISION} or {SNAPSHOT}." ) ) def lookup_missing_hashes(grouped_swhids: Dict[str, List[bytes]]) -> Set[str]: """Lookup missing Software Heritage persistent identifier hash, using batch processing. Args: A dictionary with: keys: object types values: object hashes Returns: A set(hexadecimal) of the hashes not found in the storage """ missing_hashes = [] for obj_type, obj_ids in grouped_swhids.items(): if obj_type == CONTENT: missing_hashes.append(storage.content_missing_per_sha1_git(obj_ids)) elif obj_type == DIRECTORY: missing_hashes.append(storage.directory_missing(obj_ids)) elif obj_type == REVISION: missing_hashes.append(storage.revision_missing(obj_ids)) elif obj_type == RELEASE: missing_hashes.append(storage.release_missing(obj_ids)) elif obj_type == SNAPSHOT: missing_hashes.append(storage.snapshot_missing(obj_ids)) missing = set( map(lambda x: hashutil.hash_to_hex(x), itertools.chain(*missing_hashes)) ) return missing def lookup_origins_by_sha1s(sha1s: List[str]) -> Iterator[Optional[OriginInfo]]: """Lookup origins from the sha1 hash values of their URLs. Args: sha1s: list of sha1s hexadecimal representation Yields: origin information as dict """ sha1s_bytes = [hashutil.hash_to_bytes(sha1) for sha1 in sha1s] origins = storage.origin_get_by_sha1(sha1s_bytes) for origin in origins: yield converters.from_origin(origin) diff --git a/swh/web/common/identifiers.py b/swh/web/common/identifiers.py index 05a8f254..97f3643c 100644 --- a/swh/web/common/identifiers.py +++ b/swh/web/common/identifiers.py @@ -1,393 +1,381 @@ # Copyright (C) 2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from typing import Any, Dict, Iterable, List, Optional from urllib.parse import quote, unquote from typing_extensions import TypedDict from django.http import QueryDict from swh.model.exceptions import ValidationError from swh.model.hashutil import hash_to_bytes, hash_to_hex from swh.model.identifiers import ( CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT, ObjectType, QualifiedSWHID, ) from swh.web.common import archive from swh.web.common.exc import BadInputExc from swh.web.common.typing import ( QueryParameters, SnapshotContext, SWHIDContext, SWHIDInfo, SWHObjectInfo, ) from swh.web.common.utils import reverse def gen_swhid( object_type: str, object_id: str, scheme_version: int = 1, metadata: SWHIDContext = {}, ) -> str: """ Returns the SoftWare Heritage persistent IDentifier for a swh object based on: * the object type * the object id * the SWHID scheme version Args: object_type: the swh object type (content/directory/release/revision/snapshot) object_id: the swh object id (hexadecimal representation of its hash value) scheme_version: the scheme version of the SWHIDs Returns: the SWHID of the object Raises: BadInputExc: if the provided parameters do not enable to generate a valid identifier """ try: decoded_object_type = ObjectType[object_type.upper()] decoded_object_id = hash_to_bytes(object_id) obj_swhid = str( QualifiedSWHID( object_type=decoded_object_type, object_id=decoded_object_id, scheme_version=scheme_version, **metadata, ) ) except (ValidationError, KeyError, ValueError) as e: raise BadInputExc("Invalid object (%s) for SWHID. %s" % (object_id, e)) else: return obj_swhid class ResolvedSWHID(TypedDict): """parsed SWHID with context""" swhid_parsed: QualifiedSWHID """URL to browse object according to SWHID context""" browse_url: Optional[str] def resolve_swhid( swhid: str, query_params: Optional[QueryParameters] = None ) -> ResolvedSWHID: """ Try to resolve a SoftWare Heritage persistent IDentifier into an url for browsing the targeted object. Args: swhid: a SoftWare Heritage persistent IDentifier query_params: optional dict filled with query parameters to append to the browse url Returns: a dict with the following keys: * **swhid_parsed**: the parsed identifier * **browse_url**: the url for browsing the targeted object """ swhid_parsed = get_swhid(swhid) object_type = swhid_parsed.object_type object_id = swhid_parsed.object_id browse_url = None url_args = {} query_dict = QueryDict("", mutable=True) fragment = "" process_lines = object_type == ObjectType.CONTENT if query_params and len(query_params) > 0: for k in sorted(query_params.keys()): query_dict[k] = query_params[k] if swhid_parsed.origin: origin_url = unquote(swhid_parsed.origin) origin_url = archive.lookup_origin({"url": origin_url})["url"] query_dict["origin_url"] = origin_url if swhid_parsed.path and swhid_parsed.path != b"/": query_dict["path"] = swhid_parsed.path.decode("utf8", errors="replace") if swhid_parsed.anchor: directory = b"" if swhid_parsed.anchor.object_type == ObjectType.DIRECTORY: directory = swhid_parsed.anchor.object_id elif swhid_parsed.anchor.object_type == ObjectType.REVISION: revision = archive.lookup_revision( hash_to_hex(swhid_parsed.anchor.object_id) ) directory = revision["directory"] elif swhid_parsed.anchor.object_type == ObjectType.RELEASE: release = archive.lookup_release( hash_to_hex(swhid_parsed.anchor.object_id) ) if release["target_type"] == REVISION: revision = archive.lookup_revision(release["target"]) directory = revision["directory"] if object_type == ObjectType.CONTENT: if not swhid_parsed.origin: # when no origin context, content objects need to have their # path prefixed by root directory id for proper breadcrumbs display query_dict["path"] = hash_to_hex(directory) + query_dict["path"] else: # remove leading slash from SWHID content path query_dict["path"] = query_dict["path"][1:] elif object_type == ObjectType.DIRECTORY: object_id = directory # remove leading and trailing slashes from SWHID directory path if query_dict["path"].endswith("/"): query_dict["path"] = query_dict["path"][1:-1] else: query_dict["path"] = query_dict["path"][1:] # snapshot context if swhid_parsed.visit: if swhid_parsed.visit.object_type != ObjectType.SNAPSHOT: raise BadInputExc("Visit must be a snapshot SWHID.") query_dict["snapshot"] = hash_to_hex(swhid_parsed.visit.object_id) if swhid_parsed.anchor: - if swhid_parsed.anchor.object_type == ObjectType.REVISION: - # check if the anchor revision is the tip of a branch - branch_name = archive.lookup_snapshot_branch_name_from_tip_revision( - hash_to_hex(swhid_parsed.visit.object_id), - hash_to_hex(swhid_parsed.anchor.object_id), - ) - if branch_name: - query_dict["branch"] = branch_name - elif object_type != ObjectType.REVISION: - query_dict["revision"] = hash_to_hex(swhid_parsed.anchor.object_id) + if ( + swhid_parsed.anchor.object_type == ObjectType.REVISION + and object_type != ObjectType.REVISION + ): + query_dict["revision"] = hash_to_hex(swhid_parsed.anchor.object_id) elif swhid_parsed.anchor.object_type == ObjectType.RELEASE: release = archive.lookup_release( hash_to_hex(swhid_parsed.anchor.object_id) ) if release: query_dict["release"] = release["name"] - if object_type == ObjectType.REVISION and "release" not in query_dict: - branch_name = archive.lookup_snapshot_branch_name_from_tip_revision( - hash_to_hex(swhid_parsed.visit.object_id), hash_to_hex(object_id) - ) - if branch_name: - query_dict["branch"] = branch_name - # browsing content or directory without snapshot context elif ( object_type in (ObjectType.CONTENT, ObjectType.DIRECTORY) and swhid_parsed.anchor ): if swhid_parsed.anchor.object_type == ObjectType.REVISION: # anchor revision, objects are browsed from its view object_type = ObjectType.REVISION object_id = swhid_parsed.anchor.object_id elif ( object_type == ObjectType.DIRECTORY and swhid_parsed.anchor.object_type == ObjectType.DIRECTORY ): # a directory is browsed from its root object_id = swhid_parsed.anchor.object_id if object_type == ObjectType.CONTENT: url_args["query_string"] = f"sha1_git:{hash_to_hex(object_id)}" elif object_type in (ObjectType.DIRECTORY, ObjectType.RELEASE, ObjectType.REVISION): url_args["sha1_git"] = hash_to_hex(object_id) elif object_type == ObjectType.SNAPSHOT: url_args["snapshot_id"] = hash_to_hex(object_id) if swhid_parsed.lines and process_lines: lines = swhid_parsed.lines fragment += "#L" + str(lines[0]) if lines[1]: fragment += "-L" + str(lines[1]) if url_args: browse_url = ( reverse( f"browse-{object_type.name.lower()}", url_args=url_args, query_params=query_dict, ) + fragment ) return ResolvedSWHID(swhid_parsed=swhid_parsed, browse_url=browse_url) def get_swhid(swhid: str) -> QualifiedSWHID: """Check if a SWHID is valid and return it parsed. Args: swhid: a SoftWare Heritage persistent IDentifier. Raises: BadInputExc: if the provided SWHID can not be parsed. Return: A parsed SWHID. """ try: swhid_parsed = QualifiedSWHID.from_string(swhid) except ValidationError as ve: raise BadInputExc("Error when parsing identifier: %s" % " ".join(ve.messages)) else: return swhid_parsed def group_swhids(swhids: Iterable[QualifiedSWHID],) -> Dict[str, List[bytes]]: """ Groups many SoftWare Heritage persistent IDentifiers into a dictionary depending on their type. Args: swhids: an iterable of SoftWare Heritage persistent IDentifier objects Returns: A dictionary with: keys: object types values: object hashes """ swhids_by_type: Dict[str, List[bytes]] = { CONTENT: [], DIRECTORY: [], REVISION: [], RELEASE: [], SNAPSHOT: [], } for obj_swhid in swhids: obj_id = obj_swhid.object_id obj_type = obj_swhid.object_type swhids_by_type[obj_type.name.lower()].append(hash_to_bytes(obj_id)) return swhids_by_type def get_swhids_info( swh_objects: Iterable[SWHObjectInfo], snapshot_context: Optional[SnapshotContext] = None, extra_context: Optional[Dict[str, Any]] = None, ) -> List[SWHIDInfo]: """ Returns a list of dict containing info related to SWHIDs of objects. Args: swh_objects: an iterable of dict describing archived objects snapshot_context: optional dict parameter describing the snapshot in which the objects have been found extra_context: optional dict filled with extra contextual info about the objects Returns: a list of dict containing SWHIDs info """ swhids_info = [] for swh_object in swh_objects: if not swh_object["object_id"]: swhids_info.append( SWHIDInfo( object_type=swh_object["object_type"], object_id="", swhid="", swhid_url="", context={}, swhid_with_context=None, swhid_with_context_url=None, ) ) continue object_type = swh_object["object_type"] object_id = swh_object["object_id"] swhid_context: SWHIDContext = {} if snapshot_context: if snapshot_context["origin_info"] is not None: swhid_context["origin"] = quote( snapshot_context["origin_info"]["url"], safe="/?:@&" ) if object_type != SNAPSHOT: swhid_context["visit"] = gen_swhid( SNAPSHOT, snapshot_context["snapshot_id"] ) if object_type in (CONTENT, DIRECTORY): if snapshot_context["release_id"] is not None: swhid_context["anchor"] = gen_swhid( RELEASE, snapshot_context["release_id"] ) elif snapshot_context["revision_id"] is not None: swhid_context["anchor"] = gen_swhid( REVISION, snapshot_context["revision_id"] ) if object_type in (CONTENT, DIRECTORY): if ( extra_context and "revision" in extra_context and extra_context["revision"] and "anchor" not in swhid_context ): swhid_context["anchor"] = gen_swhid(REVISION, extra_context["revision"]) elif ( extra_context and "root_directory" in extra_context and extra_context["root_directory"] and "anchor" not in swhid_context and ( object_type != DIRECTORY or extra_context["root_directory"] != object_id ) ): swhid_context["anchor"] = gen_swhid( DIRECTORY, extra_context["root_directory"] ) path = None if extra_context and "path" in extra_context: path = extra_context["path"] or "/" if "filename" in extra_context and object_type == CONTENT: path += extra_context["filename"] if object_type == DIRECTORY and path == "/": path = None if path: swhid_context["path"] = quote(path, safe="/?:@&") swhid = gen_swhid(object_type, object_id) swhid_url = reverse("browse-swhid", url_args={"swhid": swhid}) swhid_with_context = None swhid_with_context_url = None if swhid_context: swhid_with_context = gen_swhid( object_type, object_id, metadata=swhid_context ) swhid_with_context_url = reverse( "browse-swhid", url_args={"swhid": swhid_with_context} ) swhids_info.append( SWHIDInfo( object_type=object_type, object_id=object_id, swhid=swhid, swhid_url=swhid_url, context=swhid_context, swhid_with_context=swhid_with_context, swhid_with_context_url=swhid_with_context_url, ) ) return swhids_info diff --git a/swh/web/tests/common/test_archive.py b/swh/web/tests/common/test_archive.py index 4ecd4434..7c1d403f 100644 --- a/swh/web/tests/common/test_archive.py +++ b/swh/web/tests/common/test_archive.py @@ -1,1201 +1,1180 @@ # Copyright (C) 2015-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from collections import defaultdict import hashlib import itertools import random from hypothesis import given import pytest from swh.model.from_disk import DentryPerms from swh.model.hashutil import hash_to_bytes, hash_to_hex from swh.model.identifiers import CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT from swh.model.model import ( Directory, DirectoryEntry, Origin, OriginVisit, Revision, Snapshot, SnapshotBranch, TargetType, ) from swh.web.common import archive from swh.web.common.exc import BadInputExc, NotFoundExc from swh.web.common.typing import OriginInfo from swh.web.tests.conftest import ctags_json_missing, fossology_missing from swh.web.tests.data import random_content, random_sha1 from swh.web.tests.strategies import ( ancestor_revisions, content, contents, contents_with_ctags, directory, empty_content, empty_directory, invalid_sha1, new_origin, new_revision, non_ancestor_revisions, origin, release, releases, revision, revision_with_submodules, revisions, sha256, snapshot, unknown_content, unknown_contents, unknown_directory, unknown_release, unknown_revision, unknown_snapshot, visit_dates, ) @given(contents()) 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 @given(contents(), unknown_contents()) 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"} @given(content()) 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 @given(content()) 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" ) @given(contents_with_ctags()) 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 == [] @given(content()) 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 @given(contents_with_ctags()) 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") @given(content()) 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(archive_data, new_origin, visit_dates): 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()) 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 @given(invalid_sha1()) 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") @given(sha256()) 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.") @given(releases()) 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_) @given(directory()) 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" ) @given(directory()) 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 @given(release()) def test_lookup_release(archive_data, release): actual_release = archive.lookup_release(release) assert actual_release == archive_data.release_get(release) @given(revision(), invalid_sha1(), sha256()) 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") @given(revision(), unknown_revision()) 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) @given(revision(), unknown_revision()) 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) @given(ancestor_revisions()) 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 @given(non_ancestor_revisions()) 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) @given(revision()) 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") @given(revision_with_submodules()) 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 @given(revision()) 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 @given(revision()) 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 @given(revision()) 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"] @given(revision()) 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 revision["decoding_failures"] == ["message"] @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_) @given(revisions()) 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_) @given(revision()) 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 @given(origin()) 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 @given(origin()) 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"]) ) @given(content()) 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 @given(empty_content()) def test_lookup_empty_content_raw(archive_data, 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"]) ) @given(content()) 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 @given(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_) @given(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 @given(empty_directory()) def test_lookup_directory_empty(empty_directory): actual_directory_ls = list(archive.lookup_directory(empty_directory)) assert actual_directory_ls == [] @given(origin()) def test_lookup_revision_by_nothing_found(origin): with pytest.raises(NotFoundExc): archive.lookup_revision_by(origin["url"], "invalid-branch-name") @given(origin()) 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 @given(origin(), 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 ) @given(origin()) 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}) @given(origin()) 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) @given(origin()) 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) @given(ancestor_revisions()) 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) @given(revision()) def test_lookup_revision_through_with_revision(revision): assert archive.lookup_revision_through( {"sha1_git": revision} ) == archive.lookup_revision(revision) @given(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" ) @given(revision()) 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"])) @given(revision()) 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 ), ) @given(content(), directory(), release(), revision(), snapshot()) def test_lookup_known_objects( archive_data, content, directory, release, revision, snapshot ): expected = archive_data.content_find(content) assert archive.lookup_object(CONTENT, content["sha1_git"]) == expected expected = archive_data.directory_get(directory) assert archive.lookup_object(DIRECTORY, directory) == expected expected = archive_data.release_get(release) assert archive.lookup_object(RELEASE, release) == expected expected = archive_data.revision_get(revision) assert archive.lookup_object(REVISION, revision) == expected expected = {**archive_data.snapshot_get(snapshot), "next_branch": None} assert archive.lookup_object(SNAPSHOT, snapshot) == expected @given( unknown_content(), unknown_directory(), unknown_release(), unknown_revision(), unknown_snapshot(), ) def test_lookup_unknown_objects( unknown_content, unknown_directory, unknown_release, unknown_revision, unknown_snapshot, ): with pytest.raises(NotFoundExc) as e: archive.lookup_object(CONTENT, unknown_content["sha1_git"]) assert e.match(r"Content.*not found") with pytest.raises(NotFoundExc) as e: archive.lookup_object(DIRECTORY, unknown_directory) assert e.match(r"Directory.*not found") with pytest.raises(NotFoundExc) as e: archive.lookup_object(RELEASE, unknown_release) assert e.match(r"Release.*not found") with pytest.raises(NotFoundExc) as e: archive.lookup_object(REVISION, unknown_revision) assert e.match(r"Revision.*not found") with pytest.raises(NotFoundExc) as e: archive.lookup_object(SNAPSHOT, unknown_snapshot) assert e.match(r"Snapshot.*not found") @given(invalid_sha1()) def test_lookup_invalid_objects(invalid_sha1): with pytest.raises(BadInputExc) as e: archive.lookup_object("foo", invalid_sha1) assert e.match("Invalid swh object type") with pytest.raises(BadInputExc) as e: archive.lookup_object(CONTENT, invalid_sha1) assert e.match("Invalid hash") with pytest.raises(BadInputExc) as e: archive.lookup_object(DIRECTORY, invalid_sha1) assert e.match("Invalid checksum") with pytest.raises(BadInputExc) as e: archive.lookup_object(RELEASE, invalid_sha1) assert e.match("Invalid checksum") with pytest.raises(BadInputExc) as e: archive.lookup_object(REVISION, invalid_sha1) assert e.match("Invalid checksum") with pytest.raises(BadInputExc) as e: archive.lookup_object(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 = { CONTENT: [hash_to_bytes(missing_cnt)], DIRECTORY: [hash_to_bytes(missing_dir)], REVISION: [hash_to_bytes(missing_rev)], RELEASE: [hash_to_bytes(missing_rel)], 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, } @given(content(), directory()) def test_lookup_missing_hashes_some_present(archive_data, content, directory): missing_rev = random_sha1() missing_rel = random_sha1() missing_snp = random_sha1() grouped_swhids = { CONTENT: [hash_to_bytes(content["sha1_git"])], DIRECTORY: [hash_to_bytes(directory)], REVISION: [hash_to_bytes(missing_rev)], RELEASE: [hash_to_bytes(missing_rel)], SNAPSHOT: [hash_to_bytes(missing_snp)], } actual_result = archive.lookup_missing_hashes(grouped_swhids) assert actual_result == {missing_rev, missing_rel, missing_snp} @given(origin()) 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(snapshot()) -def test_lookup_snapshot_branch_name_from_tip_revision(archive_data, snapshot_id): - snapshot = archive_data.snapshot_get(snapshot_id) - branches = [ - {"name": k, "revision": v["target"]} - for k, v in snapshot["branches"].items() - if v["target_type"] == "revision" - ] - branch_info = random.choice(branches) - possible_results = [ - b["name"] for b in branches if b["revision"] == branch_info["revision"] - ] - - assert ( - archive.lookup_snapshot_branch_name_from_tip_revision( - snapshot_id, branch_info["revision"] - ) - in possible_results - ) - - @given(origin(), new_origin()) def test_lookup_origins_get_by_sha1s(origin, unknown_origin): hasher = hashlib.sha1() hasher.update(origin["url"].encode("ascii")) origin_info = OriginInfo(url=origin["url"]) origin_sha1 = hasher.hexdigest() hasher = hashlib.sha1() hasher.update(unknown_origin.url.encode("ascii")) 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] @given(snapshot()) 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 @given(revision()) 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 ) @given(snapshot()) 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 @given(revision()) 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) @given(directory(), revision()) 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 ( DIRECTORY, REVISION, ): 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_identifiers.py b/swh/web/tests/common/test_identifiers.py index 723988ee..161ac9dc 100644 --- a/swh/web/tests/common/test_identifiers.py +++ b/swh/web/tests/common/test_identifiers.py @@ -1,652 +1,637 @@ # Copyright (C) 2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import random from urllib.parse import quote from hypothesis import given import pytest from swh.model.hashutil import hash_to_bytes from swh.model.identifiers import ( CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT, QualifiedSWHID, ) from swh.model.model import Origin from swh.web.browse.snapshot_context import get_snapshot_context from swh.web.common.exc import BadInputExc from swh.web.common.identifiers import ( gen_swhid, get_swhid, get_swhids_info, group_swhids, resolve_swhid, ) from swh.web.common.typing import SWHObjectInfo from swh.web.common.utils import reverse from swh.web.tests.data import random_sha1 from swh.web.tests.strategies import ( content, directory, directory_with_subdirs, origin, origin_with_multiple_visits, release, revision, snapshot, ) @given(content()) def test_gen_swhid(content): swh_object_type = CONTENT sha1_git = content["sha1_git"] expected_swhid = "swh:1:cnt:" + sha1_git assert gen_swhid(swh_object_type, sha1_git) == expected_swhid assert ( gen_swhid(swh_object_type, sha1_git, metadata={"origin": "test"}) == expected_swhid + ";origin=test" ) assert ( gen_swhid(swh_object_type, sha1_git, metadata={"origin": None}) == expected_swhid ) with pytest.raises(BadInputExc) as e: gen_swhid("foo", sha1_git) assert e.match("Invalid object") with pytest.raises(BadInputExc) as e: gen_swhid(swh_object_type, "not a valid id") assert e.match("Invalid object") @given(content(), directory(), release(), revision(), snapshot()) def test_resolve_swhid_legacy(content, directory, release, revision, snapshot): for obj_type, obj_id in ( (CONTENT, content["sha1_git"]), (DIRECTORY, directory), (RELEASE, release), (REVISION, revision), (SNAPSHOT, snapshot), ): swhid = gen_swhid(obj_type, obj_id) url_args = {} if obj_type == CONTENT: url_args["query_string"] = f"sha1_git:{obj_id}" elif obj_type == SNAPSHOT: url_args["snapshot_id"] = obj_id else: url_args["sha1_git"] = obj_id query_params = {"origin_url": "some-origin"} browse_url = reverse( f"browse-{obj_type}", url_args=url_args, query_params=query_params ) resolved_swhid = resolve_swhid(swhid, query_params) assert isinstance(resolved_swhid["swhid_parsed"], QualifiedSWHID) assert str(resolved_swhid["swhid_parsed"]) == swhid assert resolved_swhid["browse_url"] == browse_url with pytest.raises(BadInputExc, match="'ori' is not a valid ObjectType"): resolve_swhid(f"swh:1:ori:{random_sha1()}") @given(content(), directory(), release(), revision(), snapshot()) def test_get_swhid(content, directory, release, revision, snapshot): for obj_type, obj_id in ( (CONTENT, content["sha1_git"]), (DIRECTORY, directory), (RELEASE, release), (REVISION, revision), (SNAPSHOT, snapshot), ): swhid = gen_swhid(obj_type, obj_id) swh_parsed_swhid = get_swhid(swhid) assert isinstance(swh_parsed_swhid, QualifiedSWHID) assert str(swh_parsed_swhid) == swhid with pytest.raises(BadInputExc, match="Error when parsing identifier"): get_swhid("foo") @given(content(), directory(), release(), revision(), snapshot()) def test_group_swhids(content, directory, release, revision, snapshot): swhids = [] expected = {} for obj_type, obj_id in ( (CONTENT, content["sha1_git"]), (DIRECTORY, directory), (RELEASE, release), (REVISION, revision), (SNAPSHOT, snapshot), ): swhid = gen_swhid(obj_type, obj_id) swhid = get_swhid(swhid) swhids.append(swhid) expected[obj_type] = [hash_to_bytes(obj_id)] swhid_groups = group_swhids(swhids) assert swhid_groups == expected @given(directory_with_subdirs()) def test_get_swhids_info_directory_context(archive_data, directory): swhid = get_swhids_info( [SWHObjectInfo(object_type=DIRECTORY, object_id=directory)], snapshot_context=None, )[0] assert swhid["swhid_with_context"] is None # path qualifier should be discarded for a root directory swhid = get_swhids_info( [SWHObjectInfo(object_type=DIRECTORY, object_id=directory)], snapshot_context=None, extra_context={"path": "/"}, )[0] assert swhid["swhid_with_context"] is None dir_content = archive_data.directory_ls(directory) dir_subdirs = [e for e in dir_content if e["type"] == "dir"] dir_subdir = random.choice(dir_subdirs) dir_subdir_path = f'/{dir_subdir["name"]}/' dir_subdir_content = archive_data.directory_ls(dir_subdir["target"]) dir_subdir_files = [e for e in dir_subdir_content if e["type"] == "file"] swh_objects_info = [ SWHObjectInfo(object_type=DIRECTORY, object_id=dir_subdir["target"]) ] extra_context = { "root_directory": directory, "path": dir_subdir_path, } if dir_subdir_files: dir_subdir_file = random.choice(dir_subdir_files) extra_context["filename"] = dir_subdir_file["name"] swh_objects_info.append( SWHObjectInfo( object_type=CONTENT, object_id=dir_subdir_file["checksums"]["sha1_git"] ) ) swhids = get_swhids_info( swh_objects_info, snapshot_context=None, extra_context=extra_context, ) swhid_dir_parsed = get_swhid(swhids[0]["swhid_with_context"]) anchor = gen_swhid(DIRECTORY, directory) assert swhid_dir_parsed.qualifiers() == { "anchor": anchor, "path": dir_subdir_path, } if dir_subdir_files: swhid_cnt_parsed = get_swhid(swhids[1]["swhid_with_context"]) assert swhid_cnt_parsed.qualifiers() == { "anchor": anchor, "path": f'{dir_subdir_path}{dir_subdir_file["name"]}', } @given(revision()) def test_get_swhids_info_revision_context(archive_data, revision): revision_data = archive_data.revision_get(revision) directory = revision_data["directory"] dir_content = archive_data.directory_ls(directory) dir_entry = random.choice(dir_content) swh_objects = [ SWHObjectInfo(object_type=REVISION, object_id=revision), SWHObjectInfo(object_type=DIRECTORY, object_id=directory), ] extra_context = {"revision": revision, "path": "/"} if dir_entry["type"] == "file": swh_objects.append( SWHObjectInfo( object_type=CONTENT, object_id=dir_entry["checksums"]["sha1_git"] ) ) extra_context["filename"] = dir_entry["name"] swhids = get_swhids_info( swh_objects, snapshot_context=None, extra_context=extra_context, ) assert swhids[0]["context"] == {} swhid_dir_parsed = get_swhid(swhids[1]["swhid_with_context"]) anchor = gen_swhid(REVISION, revision) assert swhid_dir_parsed.qualifiers() == { "anchor": anchor, } if dir_entry["type"] == "file": swhid_cnt_parsed = get_swhid(swhids[2]["swhid_with_context"]) assert swhid_cnt_parsed.qualifiers() == { "anchor": anchor, "path": f'/{dir_entry["name"]}', } @given(origin_with_multiple_visits()) def test_get_swhids_info_origin_snapshot_context(archive_data, origin): """ Test SWHIDs with contextual info computation under a variety of origin / snapshot browsing contexts. """ visits = archive_data.origin_visit_get(origin["url"]) for visit in visits: snapshot = archive_data.snapshot_get(visit["snapshot"]) snapshot_id = snapshot["id"] branches = { k: v["target"] for k, v in snapshot["branches"].items() if v["target_type"] == "revision" } releases = { k: v["target"] for k, v in snapshot["branches"].items() if v["target_type"] == "release" } head_rev_id = archive_data.snapshot_get_head(snapshot) head_rev = archive_data.revision_get(head_rev_id) root_dir = head_rev["directory"] dir_content = archive_data.directory_ls(root_dir) dir_files = [e for e in dir_content if e["type"] == "file"] dir_file = random.choice(dir_files) revision_log = [r["id"] for r in archive_data.revision_log(head_rev_id)] branch_name = random.choice(list(branches)) release = random.choice(list(releases)) release_data = archive_data.release_get(releases[release]) release_name = release_data["name"] revision_id = random.choice(revision_log) for snp_ctx_params, anchor_info in ( ( {"snapshot_id": snapshot_id}, {"anchor_type": REVISION, "anchor_id": head_rev_id}, ), ( {"snapshot_id": snapshot_id, "branch_name": branch_name}, {"anchor_type": REVISION, "anchor_id": branches[branch_name]}, ), ( {"snapshot_id": snapshot_id, "release_name": release_name}, {"anchor_type": RELEASE, "anchor_id": releases[release]}, ), ( {"snapshot_id": snapshot_id, "revision_id": revision_id}, {"anchor_type": REVISION, "anchor_id": revision_id}, ), ( {"origin_url": origin["url"], "snapshot_id": snapshot_id}, {"anchor_type": REVISION, "anchor_id": head_rev_id}, ), ( { "origin_url": origin["url"], "snapshot_id": snapshot_id, "branch_name": branch_name, }, {"anchor_type": REVISION, "anchor_id": branches[branch_name]}, ), ( { "origin_url": origin["url"], "snapshot_id": snapshot_id, "release_name": release_name, }, {"anchor_type": RELEASE, "anchor_id": releases[release]}, ), ( { "origin_url": origin["url"], "snapshot_id": snapshot_id, "revision_id": revision_id, }, {"anchor_type": REVISION, "anchor_id": revision_id}, ), ): snapshot_context = get_snapshot_context(**snp_ctx_params) rev_id = head_rev_id if "branch_name" in snp_ctx_params: rev_id = branches[branch_name] elif "release_name" in snp_ctx_params: rev_id = release_data["target"] elif "revision_id" in snp_ctx_params: rev_id = revision_id swh_objects = [ SWHObjectInfo( object_type=CONTENT, object_id=dir_file["checksums"]["sha1_git"] ), SWHObjectInfo(object_type=DIRECTORY, object_id=root_dir), SWHObjectInfo(object_type=REVISION, object_id=rev_id), SWHObjectInfo(object_type=SNAPSHOT, object_id=snapshot_id), ] if "release_name" in snp_ctx_params: swh_objects.append( SWHObjectInfo(object_type=RELEASE, object_id=release_data["id"]) ) swhids = get_swhids_info( swh_objects, snapshot_context, extra_context={"path": "/", "filename": dir_file["name"]}, ) swhid_cnt_parsed = get_swhid(swhids[0]["swhid_with_context"]) swhid_dir_parsed = get_swhid(swhids[1]["swhid_with_context"]) swhid_rev_parsed = get_swhid(swhids[2]["swhid_with_context"]) swhid_snp_parsed = get_swhid( swhids[3]["swhid_with_context"] or swhids[3]["swhid"] ) swhid_rel_parsed = None if "release_name" in snp_ctx_params: swhid_rel_parsed = get_swhid(swhids[4]["swhid_with_context"]) anchor = gen_swhid( object_type=anchor_info["anchor_type"], object_id=anchor_info["anchor_id"], ) snapshot_swhid = gen_swhid(object_type=SNAPSHOT, object_id=snapshot_id) expected_cnt_context = { "visit": snapshot_swhid, "anchor": anchor, "path": f'/{dir_file["name"]}', } expected_dir_context = { "visit": snapshot_swhid, "anchor": anchor, } expected_rev_context = {"visit": snapshot_swhid} expected_snp_context = {} if "origin_url" in snp_ctx_params: expected_cnt_context["origin"] = origin["url"] expected_dir_context["origin"] = origin["url"] expected_rev_context["origin"] = origin["url"] expected_snp_context["origin"] = origin["url"] assert swhid_cnt_parsed.qualifiers() == expected_cnt_context assert swhid_dir_parsed.qualifiers() == expected_dir_context assert swhid_rev_parsed.qualifiers() == expected_rev_context assert swhid_snp_parsed.qualifiers() == expected_snp_context if "release_name" in snp_ctx_params: assert swhid_rel_parsed.qualifiers() == expected_rev_context @given(origin(), directory()) def test_get_swhids_info_characters_and_url_escaping(archive_data, origin, directory): snapshot_context = get_snapshot_context(origin_url=origin["url"]) snapshot_context["origin_info"]["url"] = "http://example.org/?project=abc;def%" path = "/foo;/bar%" swhid_info = get_swhids_info( [SWHObjectInfo(object_type=DIRECTORY, object_id=directory)], snapshot_context=snapshot_context, extra_context={"path": path}, )[0] # check special characters in SWHID have been escaped assert ( swhid_info["context"]["origin"] == "http://example.org/?project%3Dabc%3Bdef%25" ) assert swhid_info["context"]["path"] == "/foo%3B/bar%25" # check special characters in SWHID URL have been escaped parsed_url_swhid = QualifiedSWHID.from_string( swhid_info["swhid_with_context_url"][1:-1] ) assert ( parsed_url_swhid.qualifiers()["origin"] == "http://example.org/%3Fproject%253Dabc%253Bdef%2525" ) assert parsed_url_swhid.qualifiers()["path"] == "/foo%253B/bar%2525" @given(origin_with_multiple_visits()) def test_resolve_swhids_snapshot_context(client, archive_data, origin): visits = archive_data.origin_visit_get(origin["url"]) visit = random.choice(visits) snapshot = archive_data.snapshot_get(visit["snapshot"]) head_rev_id = archive_data.snapshot_get_head(snapshot) branch_info = None release_info = None for branch_name in sorted(snapshot["branches"]): target_type = snapshot["branches"][branch_name]["target_type"] target = snapshot["branches"][branch_name]["target"] if target_type == "revision" and branch_info is None: branch_info = {"name": branch_name, "revision": target} elif target_type == "release" and release_info is None: release_info = {"name": branch_name, "release": target} if branch_info and release_info: break release_info["name"] = archive_data.release_get(release_info["release"])["name"] directory = archive_data.revision_get(branch_info["revision"])["directory"] directory_content = archive_data.directory_ls(directory) directory_subdirs = [e for e in directory_content if e["type"] == "dir"] directory_subdir = None if directory_subdirs: directory_subdir = random.choice(directory_subdirs) directory_files = [e for e in directory_content if e["type"] == "file"] directory_file = None if directory_files: directory_file = random.choice(directory_files) random_rev_id = random.choice(archive_data.revision_log(head_rev_id))["id"] for snp_ctx_params in ( {}, {"branch_name": branch_info["name"]}, {"release_name": release_info["name"]}, {"revision_id": random_rev_id}, ): snapshot_context = get_snapshot_context( snapshot["id"], origin["url"], **snp_ctx_params ) _check_resolved_swhid_browse_url(SNAPSHOT, snapshot["id"], snapshot_context) rev = head_rev_id if "branch_name" in snp_ctx_params: rev = branch_info["revision"] if "revision_id" in snp_ctx_params: rev = random_rev_id _check_resolved_swhid_browse_url(REVISION, rev, snapshot_context) _check_resolved_swhid_browse_url( DIRECTORY, directory, snapshot_context, path="/" ) if directory_subdir: _check_resolved_swhid_browse_url( DIRECTORY, directory_subdir["target"], snapshot_context, path=f"/{directory_subdir['name']}/", ) if directory_file: _check_resolved_swhid_browse_url( CONTENT, directory_file["target"], snapshot_context, path=f"/{directory_file['name']}", ) _check_resolved_swhid_browse_url( CONTENT, directory_file["target"], snapshot_context, path=f"/{directory_file['name']}", lines="10", ) _check_resolved_swhid_browse_url( CONTENT, directory_file["target"], snapshot_context, path=f"/{directory_file['name']}", lines="10-20", ) def _check_resolved_swhid_browse_url( object_type, object_id, snapshot_context, path=None, lines=None ): snapshot_id = snapshot_context["snapshot_id"] origin_url = None if snapshot_context["origin_info"]: origin_url = snapshot_context["origin_info"]["url"] obj_context = {} query_params = {} if origin_url: obj_context["origin"] = origin_url query_params["origin_url"] = origin_url obj_context["visit"] = gen_swhid(SNAPSHOT, snapshot_id) query_params["snapshot"] = snapshot_id if object_type in (CONTENT, DIRECTORY, REVISION): if snapshot_context["release"]: obj_context["anchor"] = gen_swhid(RELEASE, snapshot_context["release_id"]) query_params["release"] = snapshot_context["release"] else: obj_context["anchor"] = gen_swhid(REVISION, snapshot_context["revision_id"]) - if ( - snapshot_context["branch"] - and snapshot_context["branch"] != snapshot_context["revision_id"] - ): - branch = snapshot_context["branch"] - if branch == "HEAD": - for b in snapshot_context["branches"]: - if ( - b["revision"] == snapshot_context["revision_id"] - and b["name"] != "HEAD" - ): - branch = b["name"] - break - - query_params["branch"] = branch - elif object_type != REVISION: + if object_type != REVISION: query_params["revision"] = snapshot_context["revision_id"] if path: obj_context["path"] = path if path != "/": if object_type == CONTENT: query_params["path"] = path[1:] else: query_params["path"] = path[1:-1] if object_type == DIRECTORY: object_id = snapshot_context["root_directory"] if lines: obj_context["lines"] = lines obj_swhid = gen_swhid(object_type, object_id, metadata=obj_context) obj_swhid_resolved = resolve_swhid(obj_swhid) url_args = {"sha1_git": object_id} if object_type == CONTENT: url_args = {"query_string": f"sha1_git:{object_id}"} elif object_type == SNAPSHOT: url_args = {"snapshot_id": object_id} expected_url = reverse( f"browse-{object_type}", url_args=url_args, query_params=query_params, ) if lines: lines_number = lines.split("-") expected_url += f"#L{lines_number[0]}" if len(lines_number) > 1: expected_url += f"-L{lines_number[1]}" assert obj_swhid_resolved["browse_url"] == expected_url @given(directory()) def test_resolve_swhid_with_escaped_chars(directory): origin = "http://example.org/?project=abc;" origin_swhid_escaped = quote(origin, safe="/?:@&") origin_swhid_url_escaped = quote(origin, safe="/:@;") swhid = gen_swhid(DIRECTORY, directory, metadata={"origin": origin_swhid_escaped}) resolved_swhid = resolve_swhid(swhid) assert resolved_swhid["swhid_parsed"].origin == origin_swhid_escaped assert origin_swhid_url_escaped in resolved_swhid["browse_url"] @given(directory_with_subdirs()) def test_resolve_directory_swhid_path_without_trailing_slash(archive_data, directory): dir_content = archive_data.directory_ls(directory) dir_subdirs = [e for e in dir_content if e["type"] == "dir"] dir_subdir = random.choice(dir_subdirs) dir_subdir_path = dir_subdir["name"] anchor = gen_swhid(DIRECTORY, directory) swhid = gen_swhid( DIRECTORY, dir_subdir["target"], metadata={"anchor": anchor, "path": "/" + dir_subdir_path}, ) resolved_swhid = resolve_swhid(swhid) browse_url = reverse( "browse-directory", url_args={"sha1_git": directory}, query_params={"path": dir_subdir_path}, ) assert resolved_swhid["browse_url"] == browse_url @given(directory()) def test_resolve_swhid_with_malformed_origin_url(archive_data, directory): origin_url = "http://example.org/project/abc" malformed_origin_url = "http:/example.org/project/abc" archive_data.origin_add([Origin(url=origin_url)]) swhid = gen_swhid(DIRECTORY, directory, metadata={"origin": malformed_origin_url}) resolved_swhid = resolve_swhid(swhid) assert origin_url in resolved_swhid["browse_url"]