diff --git a/swh/web/common/archive.py b/swh/web/common/archive.py index 6bc8899f..7b906059 100644 --- a/swh/web/common/archive.py +++ b/swh/web/common/archive.py @@ -1,1380 +1,1388 @@ # 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, 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 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, 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, 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 """ - matches = idx_storage.origin_intrinsic_metadata_search_fulltext( - conjunction=[fulltext], limit=limit - ) - matches = [match.to_dict() for match in matches] 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 = 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) -> 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) # 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, branches_from="", branches_count=1000, target_types=None ): """Return information about a snapshot, aka the list of named branches found during a specific visit of an origin. Args: snapshot_id (str): sha1 identifier of the snapshot branches_from (str): optional parameter used to skip branches whose name is lesser than it before returning them branches_count (int): optional parameter used to restrain the amount of returned branches target_types (list): 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'`) 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 ) 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, ) 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/config.py b/swh/web/config.py index f03fa55b..db568726 100644 --- a/swh/web/config.py +++ b/swh/web/config.py @@ -1,177 +1,178 @@ # Copyright (C) 2017-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information import os from typing import Any, Dict from swh.core import config from swh.indexer.storage import get_indexer_storage from swh.scheduler import get_scheduler from swh.search import get_search from swh.storage import get_storage from swh.vault import get_vault from swh.web import settings SWH_WEB_INTERNAL_SERVER_NAME = "archive.internal.softwareheritage.org" SETTINGS_DIR = os.path.dirname(settings.__file__) DEFAULT_CONFIG = { "allowed_hosts": ("list", []), "search": ( "dict", {"cls": "remote", "url": "http://127.0.0.1:5010/", "timeout": 10,}, ), "storage": ( "dict", {"cls": "remote", "url": "http://127.0.0.1:5002/", "timeout": 10,}, ), "indexer_storage": ( "dict", {"cls": "remote", "url": "http://127.0.0.1:5007/", "timeout": 1,}, ), "log_dir": ("string", "/tmp/swh/log"), "debug": ("bool", False), "serve_assets": ("bool", False), "host": ("string", "127.0.0.1"), "port": ("int", 5004), "secret_key": ("string", "development key"), # do not display code highlighting for content > 1MB "content_display_max_size": ("int", 5 * 1024 * 1024), "snapshot_content_max_size": ("int", 1000), "throttling": ( "dict", { "cache_uri": None, # production: memcached as cache (127.0.0.1:11211) # development: in-memory cache so None "scopes": { "swh_api": { "limiter_rate": {"default": "120/h"}, "exempted_networks": ["127.0.0.0/8"], }, "swh_api_origin_search": { "limiter_rate": {"default": "10/m"}, "exempted_networks": ["127.0.0.0/8"], }, "swh_vault_cooking": { "limiter_rate": {"default": "120/h", "GET": "60/m"}, "exempted_networks": ["127.0.0.0/8"], }, "swh_save_origin": { "limiter_rate": {"default": "120/h", "POST": "10/h"}, "exempted_networks": ["127.0.0.0/8"], }, "swh_api_origin_visit_latest": { "limiter_rate": {"default": "700/m"}, "exempted_networks": ["127.0.0.0/8"], }, }, }, ), "vault": ("dict", {"cls": "remote", "args": {"url": "http://127.0.0.1:5005/",}}), "scheduler": ("dict", {"cls": "remote", "url": "http://127.0.0.1:5008/"}), "development_db": ("string", os.path.join(SETTINGS_DIR, "db.sqlite3")), "test_db": ("string", os.path.join(SETTINGS_DIR, "testdb.sqlite3")), "production_db": ("string", "/var/lib/swh/web.sqlite3"), "deposit": ( "dict", { "private_api_url": "https://deposit.softwareheritage.org/1/private/", "private_api_user": "swhworker", "private_api_password": "", }, ), "coverage_count_origins": ("bool", False), "e2e_tests_mode": ("bool", False), "es_workers_index_url": ("string", ""), "history_counters_url": ( "string", "https://stats.export.softwareheritage.org/history_counters.json", ), "client_config": ("dict", {}), "keycloak": ("dict", {"server_url": "", "realm_name": ""}), "graph": ( "dict", {"server_url": "http://graph.internal.softwareheritage.org:5009/graph/"}, ), "status": ( "dict", { "server_url": "https://status.softwareheritage.org/", "json_path": "1.0/status/578e5eddcdc0cc7951000520", }, ), + "metadata_search_backend": ("string", "swh-indexer-storage"), # or "swh-search" } swhweb_config = {} # type: Dict[str, Any] def get_config(config_file="web/web"): """Read the configuration file `config_file`. If an environment variable SWH_CONFIG_FILENAME is defined, this takes precedence over the config_file parameter. In any case, update the app with parameters (secret_key, conf) and return the parsed configuration as a dict. If no configuration file is provided, return a default configuration. """ if not swhweb_config: config_filename = os.environ.get("SWH_CONFIG_FILENAME") if config_filename: config_file = config_filename cfg = config.load_named_config(config_file, DEFAULT_CONFIG) swhweb_config.update(cfg) config.prepare_folders(swhweb_config, "log_dir") if swhweb_config.get("search"): swhweb_config["search"] = get_search(**swhweb_config["search"]) else: swhweb_config["search"] = None swhweb_config["storage"] = get_storage(**swhweb_config["storage"]) swhweb_config["vault"] = get_vault(**swhweb_config["vault"]) swhweb_config["indexer_storage"] = get_indexer_storage( **swhweb_config["indexer_storage"] ) swhweb_config["scheduler"] = get_scheduler(**swhweb_config["scheduler"]) return swhweb_config def search(): """Return the current application's search. """ return get_config()["search"] def storage(): """Return the current application's storage. """ return get_config()["storage"] def vault(): """Return the current application's vault. """ return get_config()["vault"] def indexer_storage(): """Return the current application's indexer storage. """ return get_config()["indexer_storage"] def scheduler(): """Return the current application's scheduler. """ return get_config()["scheduler"] diff --git a/swh/web/tests/api/views/test_origin.py b/swh/web/tests/api/views/test_origin.py index cfa73432..eb9fd482 100644 --- a/swh/web/tests/api/views/test_origin.py +++ b/swh/web/tests/api/views/test_origin.py @@ -1,651 +1,655 @@ # 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 datetime import timedelta from hypothesis import given import pytest from requests.utils import parse_header_links from swh.indexer.storage.model import OriginIntrinsicMetadataRow from swh.model.hashutil import hash_to_bytes from swh.model.model import Origin, OriginVisit, OriginVisitStatus from swh.storage.exc import StorageAPIError, StorageDBError from swh.storage.utils import now from swh.web.api.utils import enrich_origin, enrich_origin_visit from swh.web.common.exc import BadInputExc from swh.web.common.origin_visits import get_origin_visits from swh.web.common.utils import reverse from swh.web.tests.data import ( INDEXER_TOOL, ORIGIN_MASTER_REVISION, ORIGIN_METADATA_KEY, ORIGIN_METADATA_VALUE, ) from swh.web.tests.strategies import new_origin, new_snapshots, origin, visit_dates from swh.web.tests.utils import check_api_get_responses def _scroll_results(api_client, url): """Iterates through pages of results, and returns them all.""" results = [] while True: rv = check_api_get_responses(api_client, url, status_code=200) results.extend(rv.data) if "Link" in rv: for link in parse_header_links(rv["Link"]): if link["rel"] == "next": # Found link to next page of results url = link["url"] break else: # No link with 'rel=next' break else: # No Link header break return results def test_api_lookup_origin_visits_raise_error(api_client, mocker): mock_get_origin_visits = mocker.patch("swh.web.api.views.origin.get_origin_visits") err_msg = "voluntary error to check the bad request middleware." mock_get_origin_visits.side_effect = BadInputExc(err_msg) url = reverse("api-1-origin-visits", url_args={"origin_url": "http://foo"}) rv = check_api_get_responses(api_client, url, status_code=400) assert rv.data == {"exception": "BadInputExc", "reason": err_msg} def test_api_lookup_origin_visits_raise_swh_storage_error_db(api_client, mocker): mock_get_origin_visits = mocker.patch("swh.web.api.views.origin.get_origin_visits") err_msg = "Storage exploded! Will be back online shortly!" mock_get_origin_visits.side_effect = StorageDBError(err_msg) url = reverse("api-1-origin-visits", url_args={"origin_url": "http://foo"}) rv = check_api_get_responses(api_client, url, status_code=503) assert rv.data == { "exception": "StorageDBError", "reason": "An unexpected error occurred in the backend: %s" % err_msg, } def test_api_lookup_origin_visits_raise_swh_storage_error_api(api_client, mocker): mock_get_origin_visits = mocker.patch("swh.web.api.views.origin.get_origin_visits") err_msg = "Storage API dropped dead! Will resurrect asap!" mock_get_origin_visits.side_effect = StorageAPIError(err_msg) url = reverse("api-1-origin-visits", url_args={"origin_url": "http://foo"}) rv = check_api_get_responses(api_client, url, status_code=503) assert rv.data == { "exception": "StorageAPIError", "reason": "An unexpected error occurred in the api backend: %s" % err_msg, } @given(new_origin(), visit_dates(3), new_snapshots(3)) def test_api_lookup_origin_visits( api_client, archive_data, new_origin, visit_dates, new_snapshots ): archive_data.origin_add([new_origin]) for i, visit_date in enumerate(visit_dates): origin_visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=visit_date, type="git",)] )[0] archive_data.snapshot_add([new_snapshots[i]]) visit_status = OriginVisitStatus( origin=new_origin.url, visit=origin_visit.visit, date=now(), status="full", snapshot=new_snapshots[i].id, ) archive_data.origin_visit_status_add([visit_status]) all_visits = list(reversed(get_origin_visits(new_origin.to_dict()))) for last_visit, expected_visits in ( (None, all_visits[:2]), (all_visits[1]["visit"], all_visits[2:]), ): url = reverse( "api-1-origin-visits", url_args={"origin_url": new_origin.url}, query_params={"per_page": 2, "last_visit": last_visit}, ) rv = check_api_get_responses(api_client, url, status_code=200) for i in range(len(expected_visits)): expected_visits[i] = enrich_origin_visit( expected_visits[i], with_origin_link=False, with_origin_visit_link=True, request=rv.wsgi_request, ) assert rv.data == expected_visits @given(new_origin(), visit_dates(3), new_snapshots(3)) def test_api_lookup_origin_visits_by_id( api_client, archive_data, new_origin, visit_dates, new_snapshots ): archive_data.origin_add([new_origin]) for i, visit_date in enumerate(visit_dates): origin_visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=visit_date, type="git",)] )[0] archive_data.snapshot_add([new_snapshots[i]]) visit_status = OriginVisitStatus( origin=new_origin.url, visit=origin_visit.visit, date=now(), status="full", snapshot=new_snapshots[i].id, ) archive_data.origin_visit_status_add([visit_status]) all_visits = list(reversed(get_origin_visits(new_origin.to_dict()))) for last_visit, expected_visits in ( (None, all_visits[:2]), (all_visits[1]["visit"], all_visits[2:4]), ): url = reverse( "api-1-origin-visits", url_args={"origin_url": new_origin.url}, query_params={"per_page": 2, "last_visit": last_visit}, ) rv = check_api_get_responses(api_client, url, status_code=200) for i in range(len(expected_visits)): expected_visits[i] = enrich_origin_visit( expected_visits[i], with_origin_link=False, with_origin_visit_link=True, request=rv.wsgi_request, ) assert rv.data == expected_visits @given(new_origin(), visit_dates(3), new_snapshots(3)) def test_api_lookup_origin_visit( api_client, archive_data, new_origin, visit_dates, new_snapshots ): archive_data.origin_add([new_origin]) for i, visit_date in enumerate(visit_dates): origin_visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=visit_date, type="git",)] )[0] visit_id = origin_visit.visit archive_data.snapshot_add([new_snapshots[i]]) visit_status = OriginVisitStatus( origin=new_origin.url, visit=origin_visit.visit, date=visit_date + timedelta(minutes=5), status="full", snapshot=new_snapshots[i].id, ) archive_data.origin_visit_status_add([visit_status]) url = reverse( "api-1-origin-visit", url_args={"origin_url": new_origin.url, "visit_id": visit_id}, ) rv = check_api_get_responses(api_client, url, status_code=200) expected_visit = archive_data.origin_visit_get_by(new_origin.url, visit_id) expected_visit = enrich_origin_visit( expected_visit, with_origin_link=True, with_origin_visit_link=False, request=rv.wsgi_request, ) assert rv.data == expected_visit @given(new_origin()) def test_api_lookup_origin_visit_latest_no_visit(api_client, archive_data, new_origin): archive_data.origin_add([new_origin]) url = reverse("api-1-origin-visit-latest", url_args={"origin_url": new_origin.url}) rv = check_api_get_responses(api_client, url, status_code=404) assert rv.data == { "exception": "NotFoundExc", "reason": "No visit for origin %s found" % new_origin.url, } @given(new_origin(), visit_dates(2), new_snapshots(1)) def test_api_lookup_origin_visit_latest( api_client, archive_data, new_origin, visit_dates, new_snapshots ): archive_data.origin_add([new_origin]) visit_dates.sort() visit_ids = [] for i, visit_date in enumerate(visit_dates): origin_visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=visit_date, type="git",)] )[0] visit_ids.append(origin_visit.visit) archive_data.snapshot_add([new_snapshots[0]]) visit_status = OriginVisitStatus( origin=new_origin.url, visit=visit_ids[0], date=now(), status="full", snapshot=new_snapshots[0].id, ) archive_data.origin_visit_status_add([visit_status]) url = reverse("api-1-origin-visit-latest", url_args={"origin_url": new_origin.url}) rv = check_api_get_responses(api_client, url, status_code=200) expected_visit = archive_data.origin_visit_get_by(new_origin.url, visit_ids[1]) expected_visit = enrich_origin_visit( expected_visit, with_origin_link=True, with_origin_visit_link=False, request=rv.wsgi_request, ) assert rv.data == expected_visit @given(new_origin(), visit_dates(2), new_snapshots(1)) def test_api_lookup_origin_visit_latest_with_snapshot( api_client, archive_data, new_origin, visit_dates, new_snapshots ): archive_data.origin_add([new_origin]) visit_dates.sort() visit_ids = [] for i, visit_date in enumerate(visit_dates): origin_visit = archive_data.origin_visit_add( [OriginVisit(origin=new_origin.url, date=visit_date, type="git",)] )[0] visit_ids.append(origin_visit.visit) archive_data.snapshot_add([new_snapshots[0]]) # Add snapshot to the latest visit visit_id = visit_ids[-1] visit_status = OriginVisitStatus( origin=new_origin.url, visit=visit_id, date=now(), status="full", snapshot=new_snapshots[0].id, ) archive_data.origin_visit_status_add([visit_status]) url = reverse( "api-1-origin-visit-latest", url_args={"origin_url": new_origin.url}, query_params={"require_snapshot": True}, ) rv = check_api_get_responses(api_client, url, status_code=200) expected_visit = archive_data.origin_visit_status_get_latest( new_origin.url, type="git", require_snapshot=True ) expected_visit = enrich_origin_visit( expected_visit, with_origin_link=True, with_origin_visit_link=False, request=rv.wsgi_request, ) assert rv.data == expected_visit @given(origin()) def test_api_lookup_origin_visit_not_found(api_client, origin): all_visits = list(reversed(get_origin_visits(origin))) max_visit_id = max([v["visit"] for v in all_visits]) url = reverse( "api-1-origin-visit", url_args={"origin_url": origin["url"], "visit_id": max_visit_id + 1}, ) rv = check_api_get_responses(api_client, url, status_code=404) assert rv.data == { "exception": "NotFoundExc", "reason": "Origin %s or its visit with id %s not found!" % (origin["url"], max_visit_id + 1), } def test_api_origins_wrong_input(api_client, archive_data): """Should fail with 400 if the input is deprecated. """ # fail if wrong input url = reverse("api-1-origins", query_params={"origin_from": 1}) rv = check_api_get_responses(api_client, url, status_code=400) assert rv.data == { "exception": "BadInputExc", "reason": "Please use the Link header to browse through result", } def test_api_origins(api_client, archive_data): page_result = archive_data.origin_list(limit=10000) origins = page_result.results origin_urls = {origin.url for origin in origins} # Get only one url = reverse("api-1-origins", query_params={"origin_count": 1}) rv = check_api_get_responses(api_client, url, status_code=200) assert len(rv.data) == 1 assert {origin["url"] for origin in rv.data} <= origin_urls # Get all url = reverse("api-1-origins", query_params={"origin_count": len(origins)}) rv = check_api_get_responses(api_client, url, status_code=200) assert len(rv.data) == len(origins) assert {origin["url"] for origin in rv.data} == origin_urls # Get "all + 10" url = reverse("api-1-origins", query_params={"origin_count": len(origins) + 10}) rv = check_api_get_responses(api_client, url, status_code=200) assert len(rv.data) == len(origins) assert {origin["url"] for origin in rv.data} == origin_urls @pytest.mark.parametrize("origin_count", [1, 2, 10, 100]) def test_api_origins_scroll(api_client, archive_data, origin_count): page_result = archive_data.origin_list(limit=10000) origins = page_result.results origin_urls = {origin.url for origin in origins} url = reverse("api-1-origins", query_params={"origin_count": origin_count}) results = _scroll_results(api_client, url) assert len(results) == len(origins) assert {origin["url"] for origin in results} == origin_urls @given(origin()) def test_api_origin_by_url(api_client, archive_data, origin): origin_url = origin["url"] url = reverse("api-1-origin", url_args={"origin_url": origin_url}) rv = check_api_get_responses(api_client, url, status_code=200) expected_origin = archive_data.origin_get([origin_url])[0] expected_origin = enrich_origin(expected_origin, rv.wsgi_request) assert rv.data == expected_origin @given(new_origin()) def test_api_origin_not_found(api_client, new_origin): url = reverse("api-1-origin", url_args={"origin_url": new_origin.url}) rv = check_api_get_responses(api_client, url, status_code=404) assert rv.data == { "exception": "NotFoundExc", "reason": "Origin with url %s not found!" % new_origin.url, } @pytest.mark.parametrize("backend", ["swh-search", "swh-storage"]) def test_api_origin_search(api_client, mocker, backend): if backend != "swh-search": # equivalent to not configuring search in the config mocker.patch("swh.web.common.archive.search", None) expected_origins = { "https://github.com/wcoder/highlightjs-line-numbers.js", "https://github.com/memononen/libtess2", } # Search for 'github.com', get only one url = reverse( "api-1-origin-search", url_args={"url_pattern": "github.com"}, query_params={"limit": 1}, ) rv = check_api_get_responses(api_client, url, status_code=200) assert len(rv.data) == 1 assert {origin["url"] for origin in rv.data} <= expected_origins assert rv.data == [ enrich_origin({"url": origin["url"]}, request=rv.wsgi_request) for origin in rv.data ] # Search for 'github.com', get all url = reverse( "api-1-origin-search", url_args={"url_pattern": "github.com"}, query_params={"limit": 2}, ) rv = check_api_get_responses(api_client, url, status_code=200) assert {origin["url"] for origin in rv.data} == expected_origins assert rv.data == [ enrich_origin({"url": origin["url"]}, request=rv.wsgi_request) for origin in rv.data ] # Search for 'github.com', get more than available url = reverse( "api-1-origin-search", url_args={"url_pattern": "github.com"}, query_params={"limit": 10}, ) rv = check_api_get_responses(api_client, url, status_code=200) assert {origin["url"] for origin in rv.data} == expected_origins assert rv.data == [ enrich_origin({"url": origin["url"]}, request=rv.wsgi_request) for origin in rv.data ] @pytest.mark.parametrize("backend", ["swh-search", "swh-storage"]) def test_api_origin_search_words(api_client, mocker, backend): if backend != "swh-search": # equivalent to not configuring search in the config mocker.patch("swh.web.common.archive.search", None) expected_origins = { "https://github.com/wcoder/highlightjs-line-numbers.js", "https://github.com/memononen/libtess2", } url = reverse( "api-1-origin-search", url_args={"url_pattern": "github com"}, query_params={"limit": 2}, ) rv = check_api_get_responses(api_client, url, status_code=200) assert {origin["url"] for origin in rv.data} == expected_origins url = reverse( "api-1-origin-search", url_args={"url_pattern": "com github"}, query_params={"limit": 2}, ) rv = check_api_get_responses(api_client, url, status_code=200) assert {origin["url"] for origin in rv.data} == expected_origins url = reverse( "api-1-origin-search", url_args={"url_pattern": "memononen libtess2"}, query_params={"limit": 2}, ) rv = check_api_get_responses(api_client, url, status_code=200) assert len(rv.data) == 1 assert {origin["url"] for origin in rv.data} == { "https://github.com/memononen/libtess2" } url = reverse( "api-1-origin-search", url_args={"url_pattern": "libtess2 memononen"}, query_params={"limit": 2}, ) rv = check_api_get_responses(api_client, url, status_code=200) assert len(rv.data) == 1 assert {origin["url"] for origin in rv.data} == { "https://github.com/memononen/libtess2" } @pytest.mark.parametrize("backend", ["swh-search", "swh-storage"]) @pytest.mark.parametrize("limit", [1, 2, 3, 10]) def test_api_origin_search_scroll(api_client, archive_data, mocker, limit, backend): if backend != "swh-search": # equivalent to not configuring search in the config mocker.patch("swh.web.common.archive.search", None) expected_origins = { "https://github.com/wcoder/highlightjs-line-numbers.js", "https://github.com/memononen/libtess2", } url = reverse( "api-1-origin-search", url_args={"url_pattern": "github.com"}, query_params={"limit": limit}, ) results = _scroll_results(api_client, url) assert {origin["url"] for origin in results} == expected_origins @pytest.mark.parametrize("backend", ["swh-search", "swh-storage"]) def test_api_origin_search_limit(api_client, archive_data, tests_data, mocker, backend): if backend == "swh-search": tests_data["search"].origin_update( [{"url": "http://foobar/{}".format(i)} for i in range(2000)] ) else: # equivalent to not configuring search in the config mocker.patch("swh.web.common.archive.search", None) archive_data.origin_add( [Origin(url="http://foobar/{}".format(i)) for i in range(2000)] ) url = reverse( "api-1-origin-search", url_args={"url_pattern": "foobar"}, query_params={"limit": 1050}, ) rv = check_api_get_responses(api_client, url, status_code=200) assert len(rv.data) == 1000 -def test_api_origin_metadata_search(api_client): +@pytest.mark.parametrize("backend", ["swh-search", "swh-indexer-storage"]) +def test_api_origin_metadata_search(api_client, mocker, backend): + + mock_config = mocker.patch("swh.web.common.archive.config") + mock_config.get_config.return_value = {"metadata_search_backend": backend} url = reverse( "api-1-origin-metadata-search", query_params={"fulltext": ORIGIN_METADATA_VALUE} ) rv = check_api_get_responses(api_client, url, status_code=200) expected_data = [ { "url": origin_url, "metadata": { "from_revision": master_rev, "tool": { "name": INDEXER_TOOL["tool_name"], "version": INDEXER_TOOL["tool_version"], "configuration": INDEXER_TOOL["tool_configuration"], "id": INDEXER_TOOL["id"], }, "metadata": {ORIGIN_METADATA_KEY: ORIGIN_METADATA_VALUE}, "mappings": [], }, } for origin_url, master_rev in ORIGIN_MASTER_REVISION.items() ] assert rv.data == expected_data def test_api_origin_metadata_search_limit(api_client, mocker): mock_idx_storage = mocker.patch("swh.web.common.archive.idx_storage") oimsft = mock_idx_storage.origin_intrinsic_metadata_search_fulltext oimsft.side_effect = lambda conjunction, limit: [ OriginIntrinsicMetadataRow( id=origin_url, from_revision=hash_to_bytes(master_rev), indexer_configuration_id=INDEXER_TOOL["id"], metadata={ORIGIN_METADATA_KEY: ORIGIN_METADATA_VALUE}, mappings=[], ) for origin_url, master_rev in ORIGIN_MASTER_REVISION.items() ] url = reverse( "api-1-origin-metadata-search", query_params={"fulltext": ORIGIN_METADATA_VALUE} ) rv = check_api_get_responses(api_client, url, status_code=200) assert len(rv.data) == len(ORIGIN_MASTER_REVISION) oimsft.assert_called_with(conjunction=[ORIGIN_METADATA_VALUE], limit=70) url = reverse( "api-1-origin-metadata-search", query_params={"fulltext": ORIGIN_METADATA_VALUE, "limit": 10}, ) rv = check_api_get_responses(api_client, url, status_code=200) assert len(rv.data) == len(ORIGIN_MASTER_REVISION) oimsft.assert_called_with(conjunction=[ORIGIN_METADATA_VALUE], limit=10) url = reverse( "api-1-origin-metadata-search", query_params={"fulltext": ORIGIN_METADATA_VALUE, "limit": 987}, ) rv = check_api_get_responses(api_client, url, status_code=200) assert len(rv.data) == len(ORIGIN_MASTER_REVISION) oimsft.assert_called_with(conjunction=[ORIGIN_METADATA_VALUE], limit=100) @given(origin()) def test_api_origin_intrinsic_metadata(api_client, origin): url = reverse( "api-origin-intrinsic-metadata", url_args={"origin_url": origin["url"]} ) rv = check_api_get_responses(api_client, url, status_code=200) expected_data = {ORIGIN_METADATA_KEY: ORIGIN_METADATA_VALUE} assert rv.data == expected_data def test_api_origin_metadata_search_invalid(api_client, mocker): mock_idx_storage = mocker.patch("swh.web.common.archive.idx_storage") url = reverse("api-1-origin-metadata-search") check_api_get_responses(api_client, url, status_code=400) mock_idx_storage.assert_not_called() diff --git a/swh/web/tests/data.py b/swh/web/tests/data.py index 859ae47e..ee046f0b 100644 --- a/swh/web/tests/data.py +++ b/swh/web/tests/data.py @@ -1,402 +1,407 @@ # Copyright (C) 2018-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU Affero General Public License version 3, or any later version # See top-level LICENSE file for more information from copy import deepcopy from datetime import timedelta import os import random import time from typing import Dict, List, Optional, Set from swh.core.config import merge_configs from swh.indexer.ctags import CtagsIndexer from swh.indexer.fossology_license import FossologyLicenseIndexer from swh.indexer.mimetype import MimetypeIndexer from swh.indexer.storage import get_indexer_storage from swh.indexer.storage.model import OriginIntrinsicMetadataRow from swh.loader.git.from_disk import GitLoaderFromArchive from swh.model.hashutil import DEFAULT_ALGORITHMS, hash_to_bytes, hash_to_hex from swh.model.model import Content, Directory, Origin, OriginVisit, OriginVisitStatus from swh.search import get_search from swh.storage.algos.dir_iterators import dir_iterator from swh.storage.algos.snapshot import snapshot_get_latest from swh.storage.interface import Sha1 from swh.storage.utils import now from swh.web import config from swh.web.browse.utils import ( _re_encode_content, get_mimetype_and_encoding_for_content, prepare_content_for_display, ) from swh.web.common import archive # Module used to initialize data that will be provided as tests input # Configuration for git loader _TEST_LOADER_CONFIG = { "storage": {"cls": "memory",}, "save_data": False, "max_content_size": 100 * 1024 * 1024, } # Base content indexer configuration _TEST_INDEXER_BASE_CONFIG = { "storage": {"cls": "memory"}, "objstorage": {"cls": "memory", "args": {},}, "indexer_storage": {"cls": "memory", "args": {},}, } def random_sha1(): return hash_to_hex(bytes(random.randint(0, 255) for _ in range(20))) def random_sha256(): return hash_to_hex(bytes(random.randint(0, 255) for _ in range(32))) def random_blake2s256(): return hash_to_hex(bytes(random.randint(0, 255) for _ in range(32))) def random_content(): return { "sha1": random_sha1(), "sha1_git": random_sha1(), "sha256": random_sha256(), "blake2s256": random_blake2s256(), } _TEST_MIMETYPE_INDEXER_CONFIG = merge_configs( _TEST_INDEXER_BASE_CONFIG, { "tools": { "name": "file", "version": "1:5.30-1+deb9u1", "configuration": {"type": "library", "debian-package": "python3-magic"}, } }, ) _TEST_LICENSE_INDEXER_CONFIG = merge_configs( _TEST_INDEXER_BASE_CONFIG, { "workdir": "/tmp/swh/indexer.fossology.license", "tools": { "name": "nomos", "version": "3.1.0rc2-31-ga2cbb8c", "configuration": {"command_line": "nomossa ",}, }, }, ) _TEST_CTAGS_INDEXER_CONFIG = merge_configs( _TEST_INDEXER_BASE_CONFIG, { "workdir": "/tmp/swh/indexer.ctags", "languages": {"c": "c"}, "tools": { "name": "universal-ctags", "version": "~git7859817b", "configuration": { "command_line": """ctags --fields=+lnz --sort=no --links=no """ """--output-format=json """ }, }, }, ) # Lightweight git repositories that will be loaded to generate # input data for tests _TEST_ORIGINS = [ { "type": "git", "url": "https://github.com/memononen/libtess2", "archives": ["libtess2.zip"], }, { "type": "git", "url": "https://github.com/wcoder/highlightjs-line-numbers.js", "archives": [ "highlightjs-line-numbers.js.zip", "highlightjs-line-numbers.js_visit2.zip", ], }, { "type": "git", "url": "repo_with_submodules", "archives": ["repo_with_submodules.tgz"], }, ] _contents = {} def _add_extra_contents(storage, contents): pbm_image_data = b"""P1 # PBM example 24 7 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 1 1 1 1 0 0 1 1 1 1 0 0 1 1 1 1 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 1 0 0 1 1 1 0 0 0 1 1 1 0 0 0 1 1 1 0 0 0 1 1 1 1 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 1 1 1 0 0 1 1 1 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0""" # add file with mimetype image/x-portable-bitmap in the archive content pbm_content = Content.from_data(pbm_image_data) storage.content_add([pbm_content]) contents.add(pbm_content.sha1) INDEXER_TOOL = { "tool_name": "swh-web tests", "tool_version": "1.0", "tool_configuration": {}, } ORIGIN_METADATA_KEY = "vcs" ORIGIN_METADATA_VALUE = "git" ORIGIN_MASTER_REVISION = {} # Tests data initialization def _init_tests_data(): # To hold reference to the memory storage storage = None # Create search instance search = get_search("memory") search.initialize() search.origin_update({"url": origin["url"]} for origin in _TEST_ORIGINS) # Create indexer storage instance that will be shared by indexers idx_storage = get_indexer_storage("memory") # Declare a test tool for origin intrinsic metadata tests idx_tool = idx_storage.indexer_configuration_add([INDEXER_TOOL])[0] INDEXER_TOOL["id"] = idx_tool["id"] # Load git repositories from archives for origin in _TEST_ORIGINS: for i, archive_ in enumerate(origin["archives"]): if i > 0: # ensure visit dates will be different when simulating # multiple visits of an origin time.sleep(1) origin_repo_archive = os.path.join( os.path.dirname(__file__), "resources/repos/%s" % archive_ ) loader = GitLoaderFromArchive( origin["url"], archive_path=origin_repo_archive, config=_TEST_LOADER_CONFIG, ) if storage is None: storage = loader.storage else: loader.storage = storage result = loader.load() assert result["status"] == "eventful" ori = storage.origin_get([origin["url"]])[0] origin.update(ori.to_dict()) # add an 'id' key if enabled search.origin_update([{"url": origin["url"], "has_visits": True}]) for i in range(250): url = "https://many.origins/%d" % (i + 1) # storage.origin_add([{'url': url}]) storage.origin_add([Origin(url=url)]) search.origin_update([{"url": url, "has_visits": True}]) date = now() visit = OriginVisit(origin=url, date=date, type="tar") visit = storage.origin_visit_add([visit])[0] visit_status = OriginVisitStatus( origin=url, visit=visit.visit, date=date + timedelta(minutes=1), type=visit.type, status="full", snapshot=hash_to_bytes("1a8893e6a86f444e8be8e7bda6cb34fb1735a00e"), ) storage.origin_visit_status_add([visit_status]) sha1s: Set[Sha1] = set() directories = set() revisions = set() releases = set() snapshots = set() content_path = {} # Get all objects loaded into the test archive + metadata = {ORIGIN_METADATA_KEY: ORIGIN_METADATA_VALUE} for origin in _TEST_ORIGINS: snp = snapshot_get_latest(storage, origin["url"]) snapshots.add(hash_to_hex(snp.id)) for branch_name, branch_data in snp.branches.items(): target_type = branch_data.target_type.value if target_type == "revision": revisions.add(branch_data.target) if b"master" in branch_name: # Add some origin intrinsic metadata for tests origin_metadata = OriginIntrinsicMetadataRow( id=origin["url"], from_revision=branch_data.target, indexer_configuration_id=idx_tool["id"], - metadata={ORIGIN_METADATA_KEY: ORIGIN_METADATA_VALUE}, + metadata=metadata, mappings=[], ) idx_storage.origin_intrinsic_metadata_add([origin_metadata]) + search.origin_update( + [{"url": origin["url"], "intrinsic_metadata": metadata}] + ) + ORIGIN_MASTER_REVISION[origin["url"]] = hash_to_hex( branch_data.target ) elif target_type == "release": release = storage.release_get([branch_data.target])[0] revisions.add(release.target) releases.add(hash_to_hex(branch_data.target)) for rev_log in storage.revision_shortlog(set(revisions)): rev_id = rev_log[0] revisions.add(rev_id) for rev in storage.revision_get(revisions): if rev is None: continue dir_id = rev.directory directories.add(hash_to_hex(dir_id)) for entry in dir_iterator(storage, dir_id): if entry["type"] == "file": sha1s.add(entry["sha1"]) content_path[entry["sha1"]] = "/".join( [hash_to_hex(dir_id), entry["path"].decode("utf-8")] ) elif entry["type"] == "dir": directories.add(hash_to_hex(entry["target"])) _add_extra_contents(storage, sha1s) # Get all checksums for each content result: List[Optional[Content]] = storage.content_get(list(sha1s)) contents: List[Dict] = [] for content in result: assert content is not None sha1 = hash_to_hex(content.sha1) content_metadata = { algo: hash_to_hex(getattr(content, algo)) for algo in DEFAULT_ALGORITHMS } path = "" if content.sha1 in content_path: path = content_path[content.sha1] cnt_data = storage.content_get_data(content.sha1) assert cnt_data is not None mimetype, encoding = get_mimetype_and_encoding_for_content(cnt_data) _, _, cnt_data = _re_encode_content(mimetype, encoding, cnt_data) content_display_data = prepare_content_for_display(cnt_data, mimetype, path) content_metadata.update( { "path": path, "mimetype": mimetype, "encoding": encoding, "hljs_language": content_display_data["language"], "data": content_display_data["content_data"], } ) _contents[sha1] = content_metadata contents.append(content_metadata) # Add the empty directory to the test archive storage.directory_add([Directory(entries=())]) # Add empty content to the test archive storage.content_add([Content.from_data(data=b"")]) # Return tests data return { "search": search, "storage": storage, "idx_storage": idx_storage, "origins": _TEST_ORIGINS, "contents": contents, "directories": list(directories), "releases": list(releases), "revisions": list(map(hash_to_hex, revisions)), "snapshots": list(snapshots), "generated_checksums": set(), } def _init_indexers(tests_data): # Instantiate content indexers that will be used in tests # and force them to use the memory storages indexers = {} for idx_name, idx_class, idx_config in ( ("mimetype_indexer", MimetypeIndexer, _TEST_MIMETYPE_INDEXER_CONFIG), ("license_indexer", FossologyLicenseIndexer, _TEST_LICENSE_INDEXER_CONFIG), ("ctags_indexer", CtagsIndexer, _TEST_CTAGS_INDEXER_CONFIG), ): idx = idx_class(config=idx_config) idx.storage = tests_data["storage"] idx.objstorage = tests_data["storage"].objstorage idx.idx_storage = tests_data["idx_storage"] idx.register_tools(idx.config["tools"]) indexers[idx_name] = idx return indexers def get_content(content_sha1): return _contents.get(content_sha1) _tests_data = None _current_tests_data = None _indexer_loggers = {} def get_tests_data(reset=False): """ Initialize tests data and return them in a dict. """ global _tests_data, _current_tests_data if _tests_data is None: _tests_data = _init_tests_data() indexers = _init_indexers(_tests_data) for (name, idx) in indexers.items(): # pytest makes the loggers use a temporary file; and deepcopy # requires serializability. So we remove them, and add them # back after the copy. _indexer_loggers[name] = idx.log del idx.log _tests_data.update(indexers) if reset or _current_tests_data is None: _current_tests_data = deepcopy(_tests_data) for (name, logger) in _indexer_loggers.items(): _current_tests_data[name].log = logger return _current_tests_data def override_storages(storage, idx_storage, search): """ Helper function to replace the storages from which archive data are fetched. """ swh_config = config.get_config() swh_config.update( {"storage": storage, "indexer_storage": idx_storage, "search": search,} ) archive.storage = storage archive.idx_storage = idx_storage archive.search = search