Page MenuHomeSoftware Heritage
Files F9312690

loader.py
No OneTemporary
Actions

loader.py
View Options

# Copyright (C) 2015-2022 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU General Public License version 3, or any later version
# See top-level LICENSE file for more information
import datetime
import hashlib
import logging
import os
from pathlib import Path
import tempfile
import time
from typing import Any, ContextManager, Dict, List, Optional, Union
from urllib.parse import urlparse
from requests.exceptions import HTTPError
import sentry_sdk
from swh.core.config import load_from_envvar
from swh.core.statsd import Statsd
from swh.core.tarball import uncompress
from swh.loader.core.metadata_fetchers import CredentialsType, get_fetchers_for_lister
from swh.loader.core.utils import nix_hashes
from swh.loader.exception import NotFound, UnsupportedChecksumComputation
from swh.loader.package.utils import download
from swh.model import from_disk
from swh.model.model import (
Content,
Directory,
Origin,
OriginVisit,
OriginVisitStatus,
RawExtrinsicMetadata,
Sha1Git,
SkippedContent,
Snapshot,
SnapshotBranch,
TargetType,
)
from swh.storage import get_storage
from swh.storage.algos.snapshot import snapshot_get_latest
from swh.storage.interface import StorageInterface
from swh.storage.utils import now
DEFAULT_CONFIG: Dict[str, Any] = {
"max_content_size": 100 * 1024 * 1024,
}
SENTRY_ORIGIN_URL_TAG_NAME = "swh.loader.origin_url"
SENTRY_VISIT_TYPE_TAG_NAME = "swh.loader.visit_type"
class BaseLoader:
"""Base class for (D)VCS loaders (e.g Svn, Git, Mercurial, ...) or PackageLoader (e.g
PyPI, Npm, CRAN, ...)
A loader retrieves origin information (git/mercurial/svn repositories, pypi/npm/...
package artifacts), ingests the contents/directories/revisions/releases/snapshot
read from those artifacts and send them to the archive through the storage backend.
The main entry point for the loader is the :func:`load` function.
2 static methods (:func:`from_config`, :func:`from_configfile`) centralizes and
eases the loader instantiation from either configuration dict or configuration file.
Some class examples:
- :class:`SvnLoader`
- :class:`GitLoader`
- :class:`PyPILoader`
- :class:`NpmLoader`
Args:
lister_name: Name of the lister which triggered this load.
If provided, the loader will try to use the forge's API to retrieve extrinsic
metadata
lister_instance_name: Name of the lister instance which triggered this load.
Must be None iff lister_name is, but it may be the empty string for listers
with a single instance.
"""
visit_type: str
origin: Origin
loaded_snapshot_id: Optional[Sha1Git]
parent_origins: Optional[List[Origin]]
"""If the given origin is a "forge fork" (ie. created with the "Fork" button
of GitHub-like forges), :meth:`build_extrinsic_origin_metadata` sets this to
a list of origins it was forked from; closest parent first."""
def __init__(
self,
storage: StorageInterface,
origin_url: str,
logging_class: Optional[str] = None,
save_data_path: Optional[str] = None,
max_content_size: Optional[int] = None,
lister_name: Optional[str] = None,
lister_instance_name: Optional[str] = None,
metadata_fetcher_credentials: CredentialsType = None,
):
if lister_name == "":
raise ValueError("lister_name must not be the empty string")
if lister_name is None and lister_instance_name is not None:
raise ValueError(
f"lister_name is None but lister_instance_name is {lister_instance_name!r}"
)
if lister_name is not None and lister_instance_name is None:
raise ValueError(
f"lister_instance_name is None but lister_name is {lister_name!r}"
)
self.storage = storage
self.origin = Origin(url=origin_url)
self.max_content_size = int(max_content_size) if max_content_size else None
self.lister_name = lister_name
self.lister_instance_name = lister_instance_name
self.metadata_fetcher_credentials = metadata_fetcher_credentials or {}
if logging_class is None:
logging_class = "%s.%s" % (
self.__class__.__module__,
self.__class__.__name__,
)
self.log = logging.getLogger(logging_class)
_log = logging.getLogger("requests.packages.urllib3.connectionpool")
_log.setLevel(logging.WARN)
sentry_sdk.set_tag(SENTRY_ORIGIN_URL_TAG_NAME, self.origin.url)
sentry_sdk.set_tag(SENTRY_VISIT_TYPE_TAG_NAME, self.visit_type)
# possibly overridden in self.prepare method
self.visit_date = datetime.datetime.now(tz=datetime.timezone.utc)
self.loaded_snapshot_id = None
if save_data_path:
path = save_data_path
os.stat(path)
if not os.access(path, os.R_OK | os.W_OK):
raise PermissionError("Permission denied: %r" % path)
self.save_data_path = save_data_path
self.parent_origins = None
self.statsd = Statsd(
namespace="swh_loader", constant_tags={"visit_type": self.visit_type}
)
@classmethod
def from_config(cls, storage: Dict[str, Any], **config: Any):
"""Instantiate a loader from a configuration dict.
This is basically a backwards-compatibility shim for the CLI.
Args:
storage: instantiation config for the storage
config: the configuration dict for the loader, with the following keys:
- credentials (optional): credentials list for the scheduler
- any other kwargs passed to the loader.
Returns:
the instantiated loader
"""
# Drop the legacy config keys which aren't used for this generation of loader.
for legacy_key in ("storage", "celery"):
config.pop(legacy_key, None)
# Instantiate the storage
storage_instance = get_storage(**storage)
return cls(storage=storage_instance, **config)
@classmethod
def from_configfile(cls, **kwargs: Any):
"""Instantiate a loader from the configuration loaded from the
SWH_CONFIG_FILENAME envvar, with potential extra keyword arguments if their
value is not None.
Args:
kwargs: kwargs passed to the loader instantiation
"""
config = dict(load_from_envvar(DEFAULT_CONFIG))
config.update({k: v for k, v in kwargs.items() if v is not None})
return cls.from_config(**config)
def save_data(self) -> None:
"""Save the data associated to the current load"""
raise NotImplementedError
def get_save_data_path(self) -> str:
"""The path to which we archive the loader's raw data"""
if not hasattr(self, "__save_data_path"):
year = str(self.visit_date.year)
assert self.origin
url = self.origin.url.encode("utf-8")
origin_url_hash = hashlib.sha1(url).hexdigest()
path = "%s/sha1:%s/%s/%s" % (
self.save_data_path,
origin_url_hash[0:2],
origin_url_hash,
year,
)
os.makedirs(path, exist_ok=True)
self.__save_data_path = path
return self.__save_data_path
def flush(self) -> Dict[str, int]:
"""Flush any potential buffered data not sent to swh-storage.
Returns the same value as :meth:`swh.storage.interface.StorageInterface.flush`.
"""
return self.storage.flush()
def cleanup(self) -> None:
"""Last step executed by the loader."""
raise NotImplementedError
def _store_origin_visit(self) -> None:
"""Store origin and visit references. Sets the self.visit references."""
assert self.origin
self.storage.origin_add([self.origin])
assert isinstance(self.visit_type, str)
self.visit = list(
self.storage.origin_visit_add(
[
OriginVisit(
origin=self.origin.url,
date=self.visit_date,
type=self.visit_type,
)
]
)
)[0]
def prepare(self) -> None:
"""Second step executed by the loader to prepare some state needed by
the loader.
Raises
NotFound exception if the origin to ingest is not found.
"""
raise NotImplementedError
def get_origin(self) -> Origin:
"""Get the origin that is currently being loaded.
self.origin should be set in :func:`prepare_origin`
Returns:
dict: an origin ready to be sent to storage by
:func:`origin_add`.
"""
assert self.origin
return self.origin
def fetch_data(self) -> bool:
"""Fetch the data from the source the loader is currently loading
(ex: git/hg/svn/... repository).
Returns:
a value that is interpreted as a boolean. If True, fetch_data needs
to be called again to complete loading.
"""
raise NotImplementedError
def process_data(self) -> bool:
"""Run any additional processing between fetching and storing the data
Returns:
a value that is interpreted as a boolean. If True, fetch_data needs
to be called again to complete loading.
Ignored if ``fetch_data`` already returned :const:`False`.
"""
return True
def store_data(self) -> None:
"""Store fetched data in the database.
Should call the :func:`maybe_load_xyz` methods, which handle the
bundles sent to storage, rather than send directly.
"""
raise NotImplementedError
def load_status(self) -> Dict[str, str]:
"""Detailed loading status.
Defaults to logging an eventful load.
Returns: a dictionary that is eventually passed back as the task's
result to the scheduler, allowing tuning of the task recurrence
mechanism.
"""
return {
"status": "eventful",
}
def post_load(self, success: bool = True) -> None:
"""Permit the loader to do some additional actions according to status
after the loading is done. The flag success indicates the
loading's status.
Defaults to doing nothing.
This is up to the implementer of this method to make sure this
does not break.
Args:
success (bool): the success status of the loading
"""
pass
def visit_status(self) -> str:
"""Detailed visit status.
Defaults to logging a full visit.
"""
return "full"
def pre_cleanup(self) -> None:
"""As a first step, will try and check for dangling data to cleanup.
This should do its best to avoid raising issues.
"""
pass
def load(self) -> Dict[str, str]:
r"""Loading logic for the loader to follow:
- Store the actual ``origin_visit`` to storage
- Call :meth:`prepare` to prepare any eventual state
- Call :meth:`get_origin` to get the origin we work with and store
- while True:
- Call :meth:`fetch_data` to fetch the data to store
- Call :meth:`process_data` to optionally run processing between
:meth:`fetch_data` and :meth:`store_data`
- Call :meth:`store_data` to store the data
- Call :meth:`cleanup` to clean up any eventual state put in place
in :meth:`prepare` method.
"""
try:
with self.statsd_timed("pre_cleanup"):
self.pre_cleanup()
except Exception:
msg = "Cleaning up dangling data failed! Continue loading."
self.log.warning(msg)
sentry_sdk.capture_exception()
self._store_origin_visit()
assert (
self.visit.visit
), "The method `_store_origin_visit` should set the visit (OriginVisit)"
self.log.info(
"Load origin '%s' with type '%s'", self.origin.url, self.visit.type
)
try:
with self.statsd_timed("build_extrinsic_origin_metadata"):
metadata = self.build_extrinsic_origin_metadata()
self.load_metadata_objects(metadata)
except Exception as e:
sentry_sdk.capture_exception(e)
# Do not fail the whole task if this is the only failure
self.log.exception(
"Failure while loading extrinsic origin metadata.",
extra={
"swh_task_args": [],
"swh_task_kwargs": {
"origin": self.origin.url,
"lister_name": self.lister_name,
"lister_instance_name": self.lister_instance_name,
},
},
)
total_time_fetch_data = 0.0
total_time_process_data = 0.0
total_time_store_data = 0.0
# Initially not a success, will be True when actually one
status = "failed"
success = False
try:
with self.statsd_timed("prepare"):
self.prepare()
while True:
t1 = time.monotonic()
more_data_to_fetch = self.fetch_data()
t2 = time.monotonic()
total_time_fetch_data += t2 - t1
more_data_to_fetch = self.process_data() and more_data_to_fetch
t3 = time.monotonic()
total_time_process_data += t3 - t2
self.store_data()
t4 = time.monotonic()
total_time_store_data += t4 - t3
if not more_data_to_fetch:
break
self.statsd_timing("fetch_data", total_time_fetch_data * 1000.0)
self.statsd_timing("process_data", total_time_process_data * 1000.0)
self.statsd_timing("store_data", total_time_store_data * 1000.0)
status = self.visit_status()
visit_status = OriginVisitStatus(
origin=self.origin.url,
visit=self.visit.visit,
type=self.visit_type,
date=now(),
status=status,
snapshot=self.loaded_snapshot_id,
)
self.storage.origin_visit_status_add([visit_status])
success = True
with self.statsd_timed(
"post_load", tags={"success": success, "status": status}
):
self.post_load()
except BaseException as e:
success = False
if isinstance(e, NotFound):
status = "not_found"
task_status = "uneventful"
else:
status = "partial" if self.loaded_snapshot_id else "failed"
task_status = "failed"
self.log.exception(
"Loading failure, updating to `%s` status",
status,
extra={
"swh_task_args": [],
"swh_task_kwargs": {
"origin": self.origin.url,
"lister_name": self.lister_name,
"lister_instance_name": self.lister_instance_name,
},
},
)
if not isinstance(e, (SystemExit, KeyboardInterrupt, NotFound)):
sentry_sdk.capture_exception()
visit_status = OriginVisitStatus(
origin=self.origin.url,
visit=self.visit.visit,
type=self.visit_type,
date=now(),
status=status,
snapshot=self.loaded_snapshot_id,
)
self.storage.origin_visit_status_add([visit_status])
with self.statsd_timed(
"post_load", tags={"success": success, "status": status}
):
self.post_load(success=success)
if not isinstance(e, Exception):
# e derives from BaseException but not Exception; this is most likely
# SystemExit or KeyboardInterrupt, so we should re-raise it.
raise
return {"status": task_status}
finally:
with self.statsd_timed(
"flush", tags={"success": success, "status": status}
):
self.flush()
with self.statsd_timed(
"cleanup", tags={"success": success, "status": status}
):
self.cleanup()
return self.load_status()
def load_metadata_objects(
self, metadata_objects: List[RawExtrinsicMetadata]
) -> None:
if not metadata_objects:
return
authorities = {mo.authority for mo in metadata_objects}
self.storage.metadata_authority_add(list(authorities))
fetchers = {mo.fetcher for mo in metadata_objects}
self.storage.metadata_fetcher_add(list(fetchers))
self.storage.raw_extrinsic_metadata_add(metadata_objects)
def build_extrinsic_origin_metadata(self) -> List[RawExtrinsicMetadata]:
"""Builds a list of full RawExtrinsicMetadata objects, using
a metadata fetcher returned by :func:`get_fetcher_classes`."""
if self.lister_name is None:
self.log.debug("lister_not provided, skipping extrinsic origin metadata")
return []
assert (
self.lister_instance_name is not None
), "lister_instance_name is None, but lister_name is not"
metadata = []
fetcher_classes = get_fetchers_for_lister(self.lister_name)
self.statsd_average("metadata_fetchers", len(fetcher_classes))
for cls in fetcher_classes:
metadata_fetcher = cls(
origin=self.origin,
lister_name=self.lister_name,
lister_instance_name=self.lister_instance_name,
credentials=self.metadata_fetcher_credentials,
)
with self.statsd_timed(
"fetch_one_metadata", tags={"fetcher": cls.FETCHER_NAME}
):
metadata.extend(metadata_fetcher.get_origin_metadata())
if self.parent_origins is None:
self.parent_origins = metadata_fetcher.get_parent_origins()
self.statsd_average(
"metadata_parent_origins",
len(self.parent_origins),
tags={"fetcher": cls.FETCHER_NAME},
)
self.statsd_average("metadata_objects", len(metadata))
return metadata
def statsd_timed(self, name: str, tags: Dict[str, Any] = {}) -> ContextManager:
"""
Wrapper for :meth:`swh.core.statsd.Statsd.timed`, which uses the standard
metric name and tags for loaders.
"""
return self.statsd.timed(
"operation_duration_seconds", tags={"operation": name, **tags}
)
def statsd_timing(self, name: str, value: float, tags: Dict[str, Any] = {}) -> None:
"""
Wrapper for :meth:`swh.core.statsd.Statsd.timing`, which uses the standard
metric name and tags for loaders.
"""
self.statsd.timing(
"operation_duration_seconds", value, tags={"operation": name, **tags}
)
def statsd_average(
self, name: str, value: Union[int, float], tags: Dict[str, Any] = {}
) -> None:
"""Increments both ``{name}_sum`` (by the ``value``) and ``{name}_count``
(by ``1``), allowing to prometheus to compute the average ``value`` over
time."""
self.statsd.increment(f"{name}_sum", value, tags=tags)
self.statsd.increment(f"{name}_count", tags=tags)
class NodeLoader(BaseLoader):
"""Common class for :class:`ContentLoader` and :class:`Directoryloader`.
The "checksums" field is a dictionary of hex hashes on the object retrieved (content
or directory). When "checksums_computation" is "standard", that means the checksums
are computed on the content of the remote file to retrieve itself (as unix cli
allows, "sha1sum", "sha256sum", ...). When "checksums_computation" is "nar", the
checks is delegated to the `nix-store --dump` command, it's actually checksums on
the content of the remote artifact retrieved. Other "checksums_computation" will
raise UnsupportedChecksumComputation
The multiple "fallback" urls received are mirror urls only used to fetch the object
if the main origin is no longer available. Those are not stored.
Ingestion is considered eventful on the first ingestion. Subsequent load of the same
object should end up being an uneventful visit (matching snapshot).
"""
def __init__(
self,
storage: StorageInterface,
url: str,
checksums: Dict[str, str],
checksums_computation: str = "standard",
fallback_urls: List[str] = None,
**kwargs,
):
super().__init__(storage, url, **kwargs)
self.snapshot: Optional[Snapshot] = None
self.checksums = checksums
self.checksums_computation = checksums_computation
if self.checksums_computation not in ("nar", "standard"):
raise UnsupportedChecksumComputation(
"Unsupported checksums computations: %s",
self.checksums_computation,
)
fallback_urls_ = fallback_urls or []
self.mirror_urls: List[str] = [self.origin.url, *fallback_urls_]
# Ensure content received matched the "standard" checksums received, this
# contains the checksums when checksum_computations is "standard", it's empty
# otherwise
self.standard_hashes = (
self.checksums if self.checksums_computation == "standard" else {}
)
self.log.debug("Loader checksums computation: %s", self.checksums_computation)
def prepare(self) -> None:
self.last_snapshot = snapshot_get_latest(self.storage, self.origin.url)
def load_status(self) -> Dict[str, Any]:
return {
"status": "uneventful"
if self.last_snapshot == self.snapshot
else "eventful"
}
def cleanup(self) -> None:
self.log.debug("cleanup")
class ContentLoader(NodeLoader):
"""Basic loader for edge case content ingestion.
The output snapshot is of the form:
.. code::
id: <bytes>
branches:
HEAD:
target_type: content
target: <content-id>
"""
visit_type = "content"
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.content: Optional[Content] = None
def fetch_data(self) -> bool:
"""Retrieve the content file as a Content Object"""
errors = []
for url in self.mirror_urls:
url_ = urlparse(url)
self.log.debug(
"prepare; origin_url=%s fallback=%s scheme=%s path=%s",
self.origin.url,
url,
url_.scheme,
url_.path,
)
try:
# FIXME: Ensure no "nar" computations is required for file
with tempfile.TemporaryDirectory() as tmpdir:
file_path, _ = download(
url, dest=tmpdir, hashes=self.standard_hashes
)
if self.checksums_computation == "nar":
# hashes are not "standard", so we need an extra check to happen
self.log.debug("Content to check nar hashes: %s", file_path)
actual_checksums = nix_hashes(
Path(file_path), self.checksums.keys()
).hexdigest()
if actual_checksums != self.checksums:
errors.append(
ValueError(
f"Checksum mismatched on <{url}>: "
f"{actual_checksums} != {self.checksums}"
)
)
self.log.debug(
"Mismatched checksums <%s>: continue on next mirror "
"url if any",
url,
)
continue
with open(file_path, "rb") as file:
self.content = Content.from_data(file.read())
except ValueError as e:
errors.append(e)
self.log.debug(
"Mismatched checksums <%s>: continue on next mirror url if any",
url,
)
continue
except HTTPError as http_error:
if http_error.response.status_code == 404:
self.log.debug(
"Not found '%s', continue on next mirror url if any", url
)
continue
else:
return False # no more data to fetch
if errors:
raise errors[0]
# If we reach this point, we did not find any proper content, consider the
# origin not found
raise NotFound(f"Unknown origin {self.origin.url}.")
def process_data(self) -> bool:
"""Build the snapshot out of the Content retrieved."""
assert self.content is not None
self.snapshot = Snapshot(
branches={
b"HEAD": SnapshotBranch(
target=self.content.sha1_git,
target_type=TargetType.CONTENT,
),
}
)
return False # no more data to process
def store_data(self) -> None:
"""Store newly retrieved Content and Snapshot."""
assert self.content is not None
self.storage.content_add([self.content])
assert self.snapshot is not None
self.storage.snapshot_add([self.snapshot])
self.loaded_snapshot_id = self.snapshot.id
def visit_status(self):
return "full" if self.content and self.snapshot is not None else "partial"
class DirectoryLoader(NodeLoader):
"""Basic loader for edge case directory ingestion (through one tarball).
The output snapshot is of the form:
.. code::
id: <bytes>
branches:
HEAD:
target_type: directory
target: <directory-id>
"""
visit_type = "directory"
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.directory: Optional[from_disk.Directory] = None
self.cnts: List[Content] = None
self.skipped_cnts: List[SkippedContent] = None
self.dirs: List[Directory] = None
def fetch_data(self) -> bool:
"""Fetch directory as a tarball amongst the self.mirror_urls.
Raises NotFound if no tarball is found
"""
errors = []
for url in self.mirror_urls:
url_ = urlparse(url)
self.log.debug(
"prepare; origin_url=%s fallback=%s scheme=%s path=%s",
self.origin.url,
url,
url_.scheme,
url_.path,
)
with tempfile.TemporaryDirectory() as tmpdir:
try:
tarball_path, extrinsic_metadata = download(
url,
tmpdir,
hashes=self.standard_hashes,
extra_request_headers={"Accept-Encoding": "identity"},
)
except ValueError as e:
errors.append(e)
self.log.debug(
"Mismatched checksums <%s>: continue on next mirror url if any",
url,
)
continue
except HTTPError as http_error:
if http_error.response.status_code == 404:
self.log.debug(
"Not found <%s>: continue on next mirror url if any", url
)
continue
directory_path = Path(tmpdir) / "src"
directory_path.mkdir(parents=True, exist_ok=True)
uncompress(tarball_path, dest=str(directory_path))
self.log.debug("uncompressed path to directory: %s", directory_path)
if self.checksums_computation == "nar":
# hashes are not "standard", so we need an extra check to happen
# on the uncompressed tarball
dir_to_check = next(directory_path.iterdir())
self.log.debug("Directory to check nar hashes: %s", dir_to_check)
actual_checksums = nix_hashes(
dir_to_check, self.checksums.keys()
).hexdigest()
if actual_checksums != self.checksums:
errors.append(
ValueError(
f"Checksum mismatched on <{url}>: "
f"{actual_checksums} != {self.checksums}"
)
)
self.log.debug(
"Mismatched checksums <%s>: continue on next mirror url if any",
url,
)
continue
self.directory = from_disk.Directory.from_disk(
path=bytes(directory_path),
max_content_length=self.max_content_size,
)
# Compute the merkle dag from the top-level directory
self.cnts, self.skipped_cnts, self.dirs = from_disk.iter_directory(
self.directory
)
if self.directory is not None:
return False # no more data to fetch
if errors:
raise errors[0]
# if we reach here, we did not find any proper tarball, so consider the origin
# not found
raise NotFound(f"Unknown origin {self.origin.url}.")
def process_data(self) -> bool:
"""Build the snapshot out of the Directory retrieved."""
assert self.directory is not None
# Build the snapshot
self.snapshot = Snapshot(
branches={
b"HEAD": SnapshotBranch(
target=self.directory.hash,
target_type=TargetType.DIRECTORY,
),
}
)
return False # no more data to process
def store_data(self) -> None:
"""Store newly retrieved Content and Snapshot."""
self.log.debug("Number of skipped contents: %s", len(self.skipped_cnts))
self.storage.skipped_content_add(self.skipped_cnts)
self.log.debug("Number of contents: %s", len(self.cnts))
self.storage.content_add(self.cnts)
self.log.debug("Number of directories: %s", len(self.dirs))
self.storage.directory_add(self.dirs)
assert self.snapshot is not None
self.storage.snapshot_add([self.snapshot])
self.loaded_snapshot_id = self.snapshot.id
def visit_status(self):
return "full" if self.directory and self.snapshot is not None else "partial"

File Metadata

Mime Type
text/x-python
Expires
Thu, Jul 3, 11:03 AM (1 w, 2 d ago)
Storage Engine
blob
Storage Format
Raw Data
Storage Handle
3238171

Event Timeline

About · Developer information · Legal