#
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.
from __future__ import annotations
import fnmatch
import inspect
import os
import re
from datetime import datetime, timedelta
from functools import cached_property
from typing import TYPE_CHECKING, Any, Callable, Sequence, cast
from airflow.configuration import conf
from airflow.providers.amazon.aws.utils import validate_execute_complete_event
if TYPE_CHECKING:
from airflow.utils.context import Context
from airflow.exceptions import AirflowException
from airflow.providers.amazon.aws.hooks.s3 import S3Hook
from airflow.providers.amazon.aws.triggers.s3 import S3KeysUnchangedTrigger, S3KeyTrigger
from airflow.sensors.base import BaseSensorOperator, poke_mode_only
[docs]class S3KeySensor(BaseSensorOperator):
"""
Waits for one or multiple keys (a file-like instance on S3) to be present in a S3 bucket.
The path is just a key/value pointer to a resource for the given S3 path.
Note: S3 does not support folders directly, and only provides key/value pairs.
.. seealso::
For more information on how to use this sensor, take a look at the guide:
:ref:`howto/sensor:S3KeySensor`
:param bucket_key: The key(s) being waited on. Supports full s3:// style url
or relative path from root level. When it's specified as a full s3://
url, please leave bucket_name as `None`
:param bucket_name: Name of the S3 bucket. Only needed when ``bucket_key``
is not provided as a full ``s3://`` url. When specified, all the keys passed to ``bucket_key``
refers to this bucket
:param wildcard_match: whether the bucket_key should be interpreted as a
Unix wildcard pattern
:param check_fn: Function that receives the list of the S3 objects with the context values,
and returns a boolean:
- ``True``: the criteria is met
- ``False``: the criteria isn't met
**Example**: Wait for any S3 object size more than 1 megabyte ::
def check_fn(files: List, **kwargs) -> bool:
return any(f.get('Size', 0) > 1048576 for f in files)
:param aws_conn_id: a reference to the s3 connection
:param verify: Whether to verify SSL certificates for S3 connection.
By default, SSL certificates are verified.
You can provide the following values:
- ``False``: do not validate SSL certificates. SSL will still be used
(unless use_ssl is False), but SSL certificates will not be
verified.
- ``path/to/cert/bundle.pem``: A filename of the CA cert bundle to uses.
You can specify this argument if you want to use a different
CA cert bundle than the one used by botocore.
:param deferrable: Run operator in the deferrable mode
:param use_regex: whether to use regex to check bucket
:param metadata_keys: List of head_object attributes to gather and send to ``check_fn``.
Acceptable values: Any top level attribute returned by s3.head_object. Specify * to return
all available attributes.
Default value: "Size".
If the requested attribute is not found, the key is still included and the value is None.
"""
[docs] template_fields: Sequence[str] = ("bucket_key", "bucket_name")
def __init__(
self,
*,
bucket_key: str | list[str],
bucket_name: str | None = None,
wildcard_match: bool = False,
check_fn: Callable[..., bool] | None = None,
aws_conn_id: str | None = "aws_default",
verify: str | bool | None = None,
deferrable: bool = conf.getboolean("operators", "default_deferrable", fallback=False),
use_regex: bool = False,
metadata_keys: list[str] | None = None,
**kwargs,
):
super().__init__(**kwargs)
self.bucket_name = bucket_name
self.bucket_key = bucket_key
self.wildcard_match = wildcard_match
self.check_fn = check_fn
self.aws_conn_id = aws_conn_id
self.verify = verify
self.deferrable = deferrable
self.use_regex = use_regex
self.metadata_keys = metadata_keys if metadata_keys else ["Size"]
def _check_key(self, key, context: Context):
bucket_name, key = S3Hook.get_s3_bucket_key(self.bucket_name, key, "bucket_name", "bucket_key")
self.log.info("Poking for key : s3://%s/%s", bucket_name, key)
"""
Set variable `files` which contains a list of dict which contains attributes defined by the user
Format: [{
'Size': int
}]
"""
if self.wildcard_match:
prefix = re.split(r"[\[*?]", key, 1)[0]
keys = self.hook.get_file_metadata(prefix, bucket_name)
key_matches = [k for k in keys if fnmatch.fnmatch(k["Key"], key)]
if not key_matches:
return False
# Reduce the set of metadata to requested attributes
files = []
for f in key_matches:
metadata = {}
if "*" in self.metadata_keys:
metadata = self.hook.head_object(f["Key"], bucket_name)
else:
for key in self.metadata_keys:
try:
metadata[key] = f[key]
except KeyError:
# supplied key might be from head_object response
self.log.info("Key %s not found in response, performing head_object", key)
metadata[key] = self.hook.head_object(f["Key"], bucket_name).get(key, None)
files.append(metadata)
elif self.use_regex:
keys = self.hook.get_file_metadata("", bucket_name)
key_matches = [k for k in keys if re.match(pattern=key, string=k["Key"])]
if not key_matches:
return False
else:
obj = self.hook.head_object(key, bucket_name)
if obj is None:
return False
metadata = {}
if "*" in self.metadata_keys:
metadata = self.hook.head_object(key, bucket_name)
else:
for key in self.metadata_keys:
# backwards compatibility with original implementation
if key == "Size":
metadata[key] = obj.get("ContentLength")
else:
metadata[key] = obj.get(key, None)
files = [metadata]
if self.check_fn is not None:
# For backwards compatibility, check if the function takes a context argument
signature = inspect.signature(self.check_fn)
if any(param.kind == inspect.Parameter.VAR_KEYWORD for param in signature.parameters.values()):
return self.check_fn(files, **context)
# Otherwise, just pass the files
return self.check_fn(files)
return True
[docs] def poke(self, context: Context):
if isinstance(self.bucket_key, str):
return self._check_key(self.bucket_key, context=context)
else:
return all(self._check_key(key, context=context) for key in self.bucket_key)
[docs] def execute(self, context: Context) -> None:
"""Airflow runs this method on the worker and defers using the trigger."""
if not self.deferrable:
super().execute(context)
else:
if not self.poke(context=context):
self._defer()
def _defer(self) -> None:
"""Check for a keys in s3 and defers using the triggerer."""
self.defer(
timeout=timedelta(seconds=self.timeout),
trigger=S3KeyTrigger(
bucket_name=cast(str, self.bucket_name),
bucket_key=self.bucket_key,
wildcard_match=self.wildcard_match,
aws_conn_id=self.aws_conn_id,
verify=self.verify,
poke_interval=self.poke_interval,
should_check_fn=bool(self.check_fn),
use_regex=self.use_regex,
),
method_name="execute_complete",
)
[docs] def execute_complete(self, context: Context, event: dict[str, Any]) -> None:
"""
Execute when the trigger fires - returns immediately.
Relies on trigger to throw an exception, otherwise it assumes execution was successful.
"""
if event["status"] == "running":
found_keys = self.check_fn(event["files"]) # type: ignore[misc]
if not found_keys:
self._defer()
elif event["status"] == "error":
raise AirflowException(event["message"])
@cached_property
[docs] def hook(self) -> S3Hook:
return S3Hook(aws_conn_id=self.aws_conn_id, verify=self.verify)
@poke_mode_only
[docs]class S3KeysUnchangedSensor(BaseSensorOperator):
"""
Return True if inactivity_period has passed with no increase in the number of objects matching prefix.
Note, this sensor will not behave correctly in reschedule mode, as the state of the listed
objects in the S3 bucket will be lost between rescheduled invocations.
.. seealso::
For more information on how to use this sensor, take a look at the guide:
:ref:`howto/sensor:S3KeysUnchangedSensor`
:param bucket_name: Name of the S3 bucket
:param prefix: The prefix being waited on. Relative path from bucket root level.
:param aws_conn_id: a reference to the s3 connection
:param verify: Whether or not to verify SSL certificates for S3 connection.
By default SSL certificates are verified.
You can provide the following values:
- ``False``: do not validate SSL certificates. SSL will still be used
(unless use_ssl is False), but SSL certificates will not be
verified.
- ``path/to/cert/bundle.pem``: A filename of the CA cert bundle to uses.
You can specify this argument if you want to use a different
CA cert bundle than the one used by botocore.
:param inactivity_period: The total seconds of inactivity to designate
keys unchanged. Note, this mechanism is not real time and
this operator may not return until a poke_interval after this period
has passed with no additional objects sensed.
:param min_objects: The minimum number of objects needed for keys unchanged
sensor to be considered valid.
:param previous_objects: The set of object ids found during the last poke.
:param allow_delete: Should this sensor consider objects being deleted
between pokes valid behavior. If true a warning message will be logged
when this happens. If false an error will be raised.
:param deferrable: Run sensor in the deferrable mode
"""
[docs] template_fields: Sequence[str] = ("bucket_name", "prefix")
def __init__(
self,
*,
bucket_name: str,
prefix: str,
aws_conn_id: str | None = "aws_default",
verify: bool | str | None = None,
inactivity_period: float = 60 * 60,
min_objects: int = 1,
previous_objects: set[str] | None = None,
allow_delete: bool = True,
deferrable: bool = conf.getboolean("operators", "default_deferrable", fallback=False),
**kwargs,
) -> None:
super().__init__(**kwargs)
self.bucket_name = bucket_name
self.prefix = prefix
if inactivity_period < 0:
raise ValueError("inactivity_period must be non-negative")
self.inactivity_period = inactivity_period
self.min_objects = min_objects
self.previous_objects = previous_objects or set()
self.inactivity_seconds = 0
self.allow_delete = allow_delete
self.deferrable = deferrable
self.aws_conn_id = aws_conn_id
self.verify = verify
self.last_activity_time: datetime | None = None
@cached_property
[docs] def hook(self):
"""Returns S3Hook."""
return S3Hook(aws_conn_id=self.aws_conn_id, verify=self.verify)
[docs] def is_keys_unchanged(self, current_objects: set[str]) -> bool:
"""
Check for new objects after the inactivity_period and update the sensor state accordingly.
:param current_objects: set of object ids in bucket during last poke.
"""
current_num_objects = len(current_objects)
if current_objects > self.previous_objects:
# When new objects arrived, reset the inactivity_seconds
# and update previous_objects for the next poke.
self.log.info(
"New objects found at %s, resetting last_activity_time.",
os.path.join(self.bucket_name, self.prefix),
)
self.log.debug("New objects: %s", current_objects - self.previous_objects)
self.last_activity_time = datetime.now()
self.inactivity_seconds = 0
self.previous_objects = current_objects
return False
if self.previous_objects - current_objects:
# During the last poke interval objects were deleted.
if self.allow_delete:
deleted_objects = self.previous_objects - current_objects
self.previous_objects = current_objects
self.last_activity_time = datetime.now()
self.log.info(
"Objects were deleted during the last poke interval. Updating the "
"file counter and resetting last_activity_time:\n%s",
deleted_objects,
)
return False
raise AirflowException(
f"Illegal behavior: objects were deleted in {os.path.join(self.bucket_name, self.prefix)} between pokes."
)
if self.last_activity_time:
self.inactivity_seconds = int((datetime.now() - self.last_activity_time).total_seconds())
else:
# Handles the first poke where last inactivity time is None.
self.last_activity_time = datetime.now()
self.inactivity_seconds = 0
if self.inactivity_seconds >= self.inactivity_period:
path = os.path.join(self.bucket_name, self.prefix)
if current_num_objects >= self.min_objects:
self.log.info(
"SUCCESS: \nSensor found %s objects at %s.\n"
"Waited at least %s seconds, with no new objects uploaded.",
current_num_objects,
path,
self.inactivity_period,
)
return True
self.log.error("FAILURE: Inactivity Period passed, not enough objects found in %s", path)
return False
return False
[docs] def poke(self, context: Context):
return self.is_keys_unchanged(set(self.hook.list_keys(self.bucket_name, prefix=self.prefix)))
[docs] def execute(self, context: Context) -> None:
"""Airflow runs this method on the worker and defers using the trigger if deferrable is True."""
if not self.deferrable:
super().execute(context)
else:
if not self.poke(context):
self.defer(
timeout=timedelta(seconds=self.timeout),
trigger=S3KeysUnchangedTrigger(
bucket_name=self.bucket_name,
prefix=self.prefix,
inactivity_period=self.inactivity_period,
min_objects=self.min_objects,
previous_objects=self.previous_objects,
inactivity_seconds=self.inactivity_seconds,
allow_delete=self.allow_delete,
aws_conn_id=self.aws_conn_id,
verify=self.verify,
last_activity_time=self.last_activity_time,
),
method_name="execute_complete",
)
[docs] def execute_complete(self, context: Context, event: dict[str, Any] | None = None) -> None:
"""
Execute when the trigger fires - returns immediately.
Relies on trigger to throw an exception, otherwise it assumes execution was successful.
"""
event = validate_execute_complete_event(event)
if event and event["status"] == "error":
raise AirflowException(event["message"])
return None