• Home
• Public Interface of Airflow
• airflow.models.dag

airflow.models.dag

Module Contents

Classes

DAG	A dag (directed acyclic graph) is a collection of tasks with directional
DagTag	A tag name per dag, to allow quick filtering in the DAG view.
DagOwnerAttributes	Table defining different owner attributes. For example, a link for an owner that will be passed as
DagModel	Table containing DAG properties
DagContext	DAG context is used to keep the current DAG when DAG is used as ContextManager.

Functions

create_timetable(interval, timezone)	Create a Timetable instance from a schedule_interval argument.
get_last_dagrun(dag_id, session[, ...])	Returns the last dag run for a dag, None if there was none.
get_dataset_triggered_next_run_info(dag_ids, *, session)	Given a list of dag_ids, get string representing how close any that are dataset triggered are
dag([dag_id, description, schedule, ...])	Python dag decorator. Wraps a function into an Airflow DAG.

Attributes

log
DEFAULT_VIEW_PRESETS
ORIENTATION_PRESETS
TAG_MAX_LEN
DagStateChangeCallback
ScheduleInterval
ScheduleIntervalArg
ScheduleArg
SLAMissCallback
DEFAULT_SCHEDULE_INTERVAL

airflow.models.dag.log

airflow.models.dag.DEFAULT_VIEW_PRESETS = ['grid', 'graph', 'duration', 'gantt', 'landing_times']

airflow.models.dag.ORIENTATION_PRESETS = ['LR', 'TB', 'RL', 'BT']

airflow.models.dag.TAG_MAX_LEN = 100

airflow.models.dag.DagStateChangeCallback

airflow.models.dag.ScheduleInterval

airflow.models.dag.ScheduleIntervalArg

airflow.models.dag.ScheduleArg

airflow.models.dag.SLAMissCallback

airflow.models.dag.DEFAULT_SCHEDULE_INTERVAL

exception airflow.models.dag.InconsistentDataInterval(instance, start_field_name, end_field_name)

Bases: airflow.exceptions.AirflowException

Exception raised when a model populates data interval fields incorrectly.

The data interval fields should either both be None (for runs scheduled prior to AIP-39), or both be datetime (for runs scheduled after AIP-39 is implemented). This is raised if exactly one of the fields is None.

__str__()

Return str(self).

airflow.models.dag.create_timetable(interval, timezone)

Create a Timetable instance from a schedule_interval argument.

airflow.models.dag.get_last_dagrun(dag_id, session, include_externally_triggered=False)

Returns the last dag run for a dag, None if there was none. Last dag run can be any type of run e.g. scheduled or backfilled. Overridden DagRuns are ignored.

airflow.models.dag.get_dataset_triggered_next_run_info(dag_ids, *, session)

Given a list of dag_ids, get string representing how close any that are dataset triggered are their next run, e.g. “1 of 2 datasets updated”

class airflow.models.dag.DAG(dag_id, description=None, schedule=NOTSET, schedule_interval=NOTSET, timetable=None, start_date=None, end_date=None, full_filepath=None, template_searchpath=None, template_undefined=jinja2.StrictUndefined, user_defined_macros=None, user_defined_filters=None, default_args=None, concurrency=None, max_active_tasks=airflow_conf.getint('core', 'max_active_tasks_per_dag'), max_active_runs=airflow_conf.getint('core', 'max_active_runs_per_dag'), dagrun_timeout=None, sla_miss_callback=None, default_view=airflow_conf.get_mandatory_value('webserver', 'dag_default_view').lower(), orientation=airflow_conf.get_mandatory_value('webserver', 'dag_orientation'), catchup=airflow_conf.getboolean('scheduler', 'catchup_by_default'), on_success_callback=None, on_failure_callback=None, doc_md=None, params=None, access_control=None, is_paused_upon_creation=None, jinja_environment_kwargs=None, render_template_as_native_obj=False, tags=None, owner_links=None, auto_register=True)

Bases: airflow.utils.log.logging_mixin.LoggingMixin

A dag (directed acyclic graph) is a collection of tasks with directional dependencies. A dag also has a schedule, a start date and an end date (optional). For each schedule, (say daily or hourly), the DAG needs to run each individual tasks as their dependencies are met. Certain tasks have the property of depending on their own past, meaning that they can’t run until their previous schedule (and upstream tasks) are completed.

DAGs essentially act as namespaces for tasks. A task_id can only be added once to a DAG.

Note that if you plan to use time zones all the dates provided should be pendulum dates. See Time zone aware DAGs.

New in version 2.4: The schedule argument to specify either time-based scheduling logic (timetable), or dataset-driven triggers.

Deprecated since version 2.4: The arguments schedule_interval and timetable. Their functionalities are merged into the new schedule argument.

Parameters

• dag_id (str) – The id of the DAG; must consist exclusively of alphanumeric characters, dashes, dots and underscores (all ASCII)

• description (str | None) – The description for the DAG to e.g. be shown on the webserver

• schedule (ScheduleArg) – Defines the rules according to which DAG runs are scheduled. Can accept cron string, timedelta object, Timetable, or list of Dataset objects. See also Customizing DAG Scheduling with Timetables.

• start_date (datetime.datetime | None) – The timestamp from which the scheduler will attempt to backfill

• end_date (datetime.datetime | None) – A date beyond which your DAG won’t run, leave to None for open-ended scheduling

• template_searchpath (str | Iterable[str] | None) – This list of folders (non-relative) defines where jinja will look for your templates. Order matters. Note that jinja/airflow includes the path of your DAG file by default

• template_undefined (type[jinja2.StrictUndefined]) – Template undefined type.

• user_defined_macros (dict | None) – a dictionary of macros that will be exposed in your jinja templates. For example, passing dict(foo='bar') to this argument allows you to {{ foo }} in all jinja templates related to this DAG. Note that you can pass any type of object here.

• user_defined_filters (dict | None) – a dictionary of filters that will be exposed in your jinja templates. For example, passing dict(hello=lambda name: 'Hello %s' % name) to this argument allows you to {{ 'world' | hello }} in all jinja templates related to this DAG.

• default_args (dict | None) – A dictionary of default parameters to be used as constructor keyword parameters when initialising operators. Note that operators have the same hook, and precede those defined here, meaning that if your dict contains ‘depends_on_past’: True here and ‘depends_on_past’: False in the operator’s call default_args, the actual value will be False.

• params (collections.abc.MutableMapping | None) – a dictionary of DAG level parameters that are made accessible in templates, namespaced under params. These params can be overridden at the task level.

• max_active_tasks (int) – the number of task instances allowed to run concurrently

• max_active_runs (int) – maximum number of active DAG runs, beyond this number of DAG runs in a running state, the scheduler won’t create new active DAG runs

• dagrun_timeout (datetime.timedelta | None) – specify how long a DagRun should be up before timing out / failing, so that new DagRuns can be created.

• sla_miss_callback (None | SLAMissCallback | list[SLAMissCallback]) – specify a function or list of functions to call when reporting SLA timeouts. See sla_miss_callback for more information about the function signature and parameters that are passed to the callback.

• default_view (str) – Specify DAG default view (grid, graph, duration, gantt, landing_times), default grid

• orientation (str) – Specify DAG orientation in graph view (LR, TB, RL, BT), default LR

• catchup (bool) – Perform scheduler catchup (or only run latest)? Defaults to True

• on_failure_callback (None | DagStateChangeCallback | list[DagStateChangeCallback]) – A function or list of functions to be called when a DagRun of this dag fails. A context dictionary is passed as a single parameter to this function.

• on_success_callback (None | DagStateChangeCallback | list[DagStateChangeCallback]) – Much like the on_failure_callback except that it is executed when the dag succeeds.

• access_control (dict | None) – Specify optional DAG-level actions, e.g., “{‘role1’: {‘can_read’}, ‘role2’: {‘can_read’, ‘can_edit’, ‘can_delete’}}”

• is_paused_upon_creation (bool | None) – Specifies if the dag is paused when created for the first time. If the dag exists already, this flag will be ignored. If this optional parameter is not specified, the global config setting will be used.

• jinja_environment_kwargs (dict | None) –

  additional configuration options to be passed to Jinja Environment for template rendering

  Example: to avoid Jinja from removing a trailing newline from template strings

  DAG(dag_id='my-dag',
      jinja_environment_kwargs={
          'keep_trailing_newline': True,
          # some other jinja2 Environment options here
      }
  )
  

  See: Jinja Environment documentation

• render_template_as_native_obj (bool) – If True, uses a Jinja NativeEnvironment to render templates as native Python types. If False, a Jinja Environment is used to render templates as string values.

• tags (list[str] | None) – List of tags to help filtering DAGs in the UI.

• owner_links (dict[str, str] | None) – Dict of owners and their links, that will be clickable on the DAGs view UI. Can be used as an HTTP link (for example the link to your Slack channel), or a mailto link. e.g: {“dag_owner”: “https://airflow.apache.org/”}

• auto_register (bool) – Automatically register this DAG when it is used in a with block

property dag_id: str

property is_subdag: bool

property concurrency: int

property max_active_tasks: int

property access_control

property description: str | None

property default_view: str

property pickle_id: int | None

property tasks: list[airflow.models.operator.Operator]

property task_ids: list[str]

property task_group: airflow.utils.task_group.TaskGroup

property relative_fileloc: pathlib.Path

File location of the importable dag ‘file’ relative to the configured DAGs folder.

property folder: str

Folder location of where the DAG object is instantiated.

property owner: str

Return list of all owners found in DAG tasks.

Returns

Comma separated list of owners in DAG tasks

Return type

str

property allow_future_exec_dates: bool

property concurrency_reached

This attribute is deprecated. Please use airflow.models.DAG.get_concurrency_reached method.

property is_paused

This attribute is deprecated. Please use airflow.models.DAG.get_is_paused method.

property normalized_schedule_interval: ScheduleInterval

property latest_execution_date

This attribute is deprecated. Please use airflow.models.DAG.get_latest_execution_date.

property subdags

Returns a list of the subdag objects associated to this DAG

property roots: list[airflow.models.operator.Operator]

Return nodes with no parents. These are first to execute and are called roots or root nodes.

property leaves: list[airflow.models.operator.Operator]

Return nodes with no children. These are last to execute and are called leaves or leaf nodes.

property task: airflow.decorators.TaskDecoratorCollection

fileloc: str

File path that needs to be imported to load this DAG or subdag.

This may not be an actual file on disk in the case when this DAG is loaded from a ZIP file or other DAG distribution format.

parent_dag: DAG | None

get_doc_md(doc_md)

validate()

Validate the DAG has a coherent setup.

This is called by the DAG bag before bagging the DAG.

__repr__()

Return repr(self).

__eq__(other)

Return self==value.

__ne__(other)

Return self!=value.

__lt__(other)

Return self<value.

__hash__()

Return hash(self).

__enter__()

__exit__(_type, _value, _tb)

date_range(start_date, num=None, end_date=None)

is_fixed_time_schedule()

following_schedule(dttm)

Calculates the following schedule for this dag in UTC.

Parameters

dttm – utc datetime

Returns

utc datetime

previous_schedule(dttm)

next_dagrun_info(last_automated_dagrun, *, restricted=True)

Get information about the next DagRun of this dag after date_last_automated_dagrun.

This calculates what time interval the next DagRun should operate on (its execution date) and when it can be scheduled, according to the dag’s timetable, start_date, end_date, etc. This doesn’t check max active run or any other “max_active_tasks” type limits, but only performs calculations based on the various date and interval fields of this dag and its tasks.

Parameters

• last_automated_dagrun (None | datetime.datetime | airflow.timetables.base.DataInterval) – The max(execution_date) of existing “automated” DagRuns for this dag (scheduled or backfill, but not manual).

• restricted (bool) – If set to False (default is True), ignore start_date, end_date, and catchup specified on the DAG or tasks.

Returns

DagRunInfo of the next dagrun, or None if a dagrun is not going to be scheduled.

Return type

airflow.timetables.base.DagRunInfo | None

next_dagrun_after_date(date_last_automated_dagrun)

iter_dagrun_infos_between(earliest, latest, *, align=True)

Yield DagRunInfo using this DAG’s timetable between given interval.

DagRunInfo instances yielded if their logical_date is not earlier than earliest, nor later than latest. The instances are ordered by their logical_date from earliest to latest.

If align is False, the first run will happen immediately on earliest, even if it does not fall on the logical timetable schedule. The default is True, but subdags will ignore this value and always behave as if this is set to False for backward compatibility.

Example: A DAG is scheduled to run every midnight (0 0 * * *). If earliest is 2021-06-03 23:00:00, the first DagRunInfo would be 2021-06-03 23:00:00 if align=False, and 2021-06-04 00:00:00 if align=True.

get_run_dates(start_date, end_date=None)

Returns a list of dates between the interval received as parameter using this dag’s schedule interval. Returned dates can be used for execution dates.

Parameters

• start_date – The start date of the interval.

• end_date – The end date of the interval. Defaults to timezone.utcnow().

Returns

A list of dates within the interval following the dag’s schedule.

Return type

list

normalize_schedule(dttm)

get_last_dagrun(session=NEW_SESSION, include_externally_triggered=False)

has_dag_runs(session=NEW_SESSION, include_externally_triggered=True)

param(name, default=NOTSET)

Return a DagParam object for current dag.

Parameters

• name (str) – dag parameter name.

• default (Any) – fallback value for dag parameter.

Returns

DagParam instance for specified name and current dag.

Return type

airflow.models.param.DagParam

get_concurrency_reached(session=NEW_SESSION)

Returns a boolean indicating whether the max_active_tasks limit for this DAG has been reached

get_is_active(session=NEW_SESSION)

Returns a boolean indicating whether this DAG is active

get_is_paused(session=NEW_SESSION)

Returns a boolean indicating whether this DAG is paused

handle_callback(dagrun, success=True, reason=None, session=NEW_SESSION)

Triggers the appropriate callback depending on the value of success, namely the on_failure_callback or on_success_callback. This method gets the context of a single TaskInstance part of this DagRun and passes that to the callable along with a ‘reason’, primarily to differentiate DagRun failures.

Parameters

• dagrun – DagRun object

• success – Flag to specify if failure or success callback should be called

• reason – Completion reason

• session – Database session

get_active_runs()

Returns a list of dag run execution dates currently running

Returns

List of execution dates

get_num_active_runs(external_trigger=None, only_running=True, session=NEW_SESSION)

Returns the number of active “running” dag runs

Parameters

• external_trigger – True for externally triggered active dag runs

• session –

Returns

number greater than 0 for active dag runs

get_dagrun(execution_date=None, run_id=None, session=NEW_SESSION)

Returns the dag run for a given execution date or run_id if it exists, otherwise none.

Parameters

• execution_date (datetime.datetime | None) – The execution date of the DagRun to find.

• run_id (str | None) – The run_id of the DagRun to find.

• session (sqlalchemy.orm.session.Session) –

Returns

The DagRun if found, otherwise None.

get_dagruns_between(start_date, end_date, session=NEW_SESSION)

Returns the list of dag runs between start_date (inclusive) and end_date (inclusive).

Parameters

• start_date – The starting execution date of the DagRun to find.

• end_date – The ending execution date of the DagRun to find.

• session –

Returns

The list of DagRuns found.

get_latest_execution_date(session=NEW_SESSION)

Returns the latest date for which at least one dag run exists

resolve_template_files()

get_template_env(*, force_sandboxed=False)

Build a Jinja2 environment.

set_dependency(upstream_task_id, downstream_task_id)

Simple utility method to set dependency between two tasks that already have been added to the DAG using add_task()

get_task_instances_before(base_date, num, *, session=NEW_SESSION)

Get num task instances before (including) base_date.

The returned list may contain exactly num task instances corresponding to any DagRunType. It can have less if there are less than num scheduled DAG runs before base_date.

get_task_instances(start_date=None, end_date=None, state=None, session=NEW_SESSION)

set_task_instance_state(*, task_id, map_indexes=None, execution_date=None, run_id=None, state, upstream=False, downstream=False, future=False, past=False, commit=True, session=NEW_SESSION)

Set the state of a TaskInstance to the given state, and clear its downstream tasks that are in failed or upstream_failed state.

Parameters

• task_id (str) – Task ID of the TaskInstance

• map_indexes (Collection[int] | None) – Only set TaskInstance if its map_index matches. If None (default), all mapped TaskInstances of the task are set.

• execution_date (datetime.datetime | None) – Execution date of the TaskInstance

• run_id (str | None) – The run_id of the TaskInstance

• state (airflow.utils.state.TaskInstanceState) – State to set the TaskInstance to

• upstream (bool) – Include all upstream tasks of the given task_id

• downstream (bool) – Include all downstream tasks of the given task_id

• future (bool) – Include all future TaskInstances of the given task_id

• commit (bool) – Commit changes

• past (bool) – Include all past TaskInstances of the given task_id

topological_sort(include_subdag_tasks=False)

Sorts tasks in topographical order, such that a task comes after any of its upstream dependencies.

Deprecated in place of task_group.topological_sort

set_dag_runs_state(state=State.RUNNING, session=NEW_SESSION, start_date=None, end_date=None, dag_ids=[])

clear(task_ids=None, start_date=None, end_date=None, only_failed=False, only_running=False, confirm_prompt=False, include_subdags=True, include_parentdag=True, dag_run_state=DagRunState.QUEUED, dry_run=False, session=NEW_SESSION, get_tis=False, recursion_depth=0, max_recursion_depth=None, dag_bag=None, exclude_task_ids=frozenset())

Clears a set of task instances associated with the current dag for a specified date range.

Parameters

• task_ids (Collection[str | tuple[str, int]] | None) – List of task ids or (task_id, map_index) tuples to clear

• start_date (datetime.datetime | None) – The minimum execution_date to clear

• end_date (datetime.datetime | None) – The maximum execution_date to clear

• only_failed (bool) – Only clear failed tasks

• only_running (bool) – Only clear running tasks.

• confirm_prompt (bool) – Ask for confirmation

• include_subdags (bool) – Clear tasks in subdags and clear external tasks indicated by ExternalTaskMarker

• include_parentdag (bool) – Clear tasks in the parent dag of the subdag.

• dag_run_state (airflow.utils.state.DagRunState) – state to set DagRun to. If set to False, dagrun state will not be changed.

• dry_run (bool) – Find the tasks to clear but don’t clear them.

• session (sqlalchemy.orm.session.Session) – The sqlalchemy session to use

• dag_bag (airflow.models.dagbag.DagBag | None) – The DagBag used to find the dags subdags (Optional)

• exclude_task_ids (frozenset[str] | frozenset[tuple[str, int]] | None) – A set of task_id or (task_id, map_index) tuples that should not be cleared

classmethod clear_dags(dags, start_date=None, end_date=None, only_failed=False, only_running=False, confirm_prompt=False, include_subdags=True, include_parentdag=False, dag_run_state=DagRunState.QUEUED, dry_run=False)

__deepcopy__(memo)

sub_dag(*args, **kwargs)

This method is deprecated in favor of partial_subset

partial_subset(task_ids_or_regex, include_downstream=False, include_upstream=True, include_direct_upstream=False)

Returns a subset of the current dag as a deep copy of the current dag based on a regex that should match one or many tasks, and includes upstream and downstream neighbours based on the flag passed.

Parameters

• task_ids_or_regex (str | Pattern | Iterable[str]) – Either a list of task_ids, or a regex to match against task ids (as a string, or compiled regex pattern).

• include_downstream – Include all downstream tasks of matched tasks, in addition to matched tasks.

• include_upstream – Include all upstream tasks of matched tasks, in addition to matched tasks.

• include_direct_upstream – Include all tasks directly upstream of matched and downstream (if include_downstream = True) tasks

has_task(task_id)

has_task_group(task_group_id)

task_group_dict()

get_task(task_id, include_subdags=False)

pickle_info()

pickle(session=NEW_SESSION)

tree_view()

Print an ASCII tree representation of the DAG.

add_task(task)

Add a task to the DAG

Parameters

task (airflow.models.operator.Operator) – the task you want to add

add_tasks(tasks)

Add a list of tasks to the DAG

Parameters

tasks (Iterable[airflow.models.operator.Operator]) – a lit of tasks you want to add

run(start_date=None, end_date=None, mark_success=False, local=False, executor=None, donot_pickle=airflow_conf.getboolean('core', 'donot_pickle'), ignore_task_deps=False, ignore_first_depends_on_past=True, pool=None, delay_on_limit_secs=1.0, verbose=False, conf=None, rerun_failed_tasks=False, run_backwards=False, run_at_least_once=False, continue_on_failures=False, disable_retry=False)

Runs the DAG.

Parameters

• start_date – the start date of the range to run

• end_date – the end date of the range to run

• mark_success – True to mark jobs as succeeded without running them

• local – True to run the tasks using the LocalExecutor

• executor – The executor instance to run the tasks

• donot_pickle – True to avoid pickling DAG object and send to workers

• ignore_task_deps – True to skip upstream tasks

• ignore_first_depends_on_past – True to ignore depends_on_past dependencies for the first set of tasks only

• pool – Resource pool to use

• delay_on_limit_secs – Time in seconds to wait before next attempt to run dag run when max_active_runs limit has been reached

• verbose – Make logging output more verbose

• conf – user defined dictionary passed from CLI

• rerun_failed_tasks –

• run_backwards –

• run_at_least_once – If true, always run the DAG at least once even if no logical run exists within the time range.

cli()

Exposes a CLI specific to this DAG

test(execution_date=None, run_conf=None, conn_file_path=None, variable_file_path=None, session=NEW_SESSION)

Execute one single DagRun for a given DAG and execution date.

Parameters

• execution_date (datetime.datetime | None) – execution date for the DAG run

• run_conf (dict[str, Any] | None) – configuration to pass to newly created dagrun

• conn_file_path (str | None) – file path to a connection file in either yaml or json

• variable_file_path (str | None) – file path to a variable file in either yaml or json

• session (sqlalchemy.orm.session.Session) – database connection (optional)

create_dagrun(state, execution_date=None, run_id=None, start_date=None, external_trigger=False, conf=None, run_type=None, session=NEW_SESSION, dag_hash=None, creating_job_id=None, data_interval=None)

Creates a dag run from this dag including the tasks associated with this dag. Returns the dag run.

Parameters

• run_id (str | None) – defines the run id for this dag run

• run_type (airflow.utils.types.DagRunType | None) – type of DagRun

• execution_date (datetime.datetime | None) – the execution date of this dag run

• state (airflow.utils.state.DagRunState) – the state of the dag run

• start_date (datetime.datetime | None) – the date this dag run should be evaluated

• external_trigger (bool | None) – whether this dag run is externally triggered

• conf (dict | None) – Dict containing configuration/parameters to pass to the DAG

• creating_job_id (int | None) – id of the job creating this DagRun

• session (sqlalchemy.orm.session.Session) – database session

• dag_hash (str | None) – Hash of Serialized DAG

• data_interval (tuple[datetime.datetime, datetime.datetime] | None) – Data interval of the DagRun

classmethod bulk_sync_to_db(dags, session=NEW_SESSION)

This method is deprecated in favor of bulk_write_to_db

classmethod bulk_write_to_db(dags, processor_subdir=None, session=NEW_SESSION)

Ensure the DagModel rows for the given dags are up-to-date in the dag table in the DB, including calculated fields.

Note that this method can be called for both DAGs and SubDAGs. A SubDag is actually a SubDagOperator.

Parameters

dags (Collection[DAG]) – the DAG objects to save to the DB

Returns

None

sync_to_db(processor_subdir=None, session=NEW_SESSION)

Save attributes about this DAG to the DB. Note that this method can be called for both DAGs and SubDAGs. A SubDag is actually a SubDagOperator.

Returns

None

get_default_view()

This is only there for backward compatible jinja2 templates

static deactivate_unknown_dags(active_dag_ids, session=NEW_SESSION)

Given a list of known DAGs, deactivate any other DAGs that are marked as active in the ORM

Parameters

active_dag_ids – list of DAG IDs that are active

Returns

None

static deactivate_stale_dags(expiration_date, session=NEW_SESSION)

Deactivate any DAGs that were last touched by the scheduler before the expiration date. These DAGs were likely deleted.

Parameters

expiration_date – set inactive DAGs that were touched before this time

Returns

None

static get_num_task_instances(dag_id, run_id=None, task_ids=None, states=None, session=NEW_SESSION)

Returns the number of task instances in the given DAG.

Parameters

• session – ORM session

• dag_id – ID of the DAG to get the task concurrency of

• run_id – ID of the DAG run to get the task concurrency of

• task_ids – A list of valid task IDs for the given DAG

• states – A list of states to filter by if supplied

Returns

The number of running tasks

Return type

int

classmethod get_serialized_fields()

Stringified DAGs and operators contain exactly these fields.

get_edge_info(upstream_task_id, downstream_task_id)

Returns edge information for the given pair of tasks if present, and an empty edge if there is no information.

set_edge_info(upstream_task_id, downstream_task_id, info)

Sets the given edge information on the DAG. Note that this will overwrite, rather than merge with, existing info.

validate_schedule_and_params()

Validates & raise exception if there are any Params in the DAG which neither have a default value nor have the null in schema[‘type’] list, but the DAG have a schedule_interval which is not None.

iter_invalid_owner_links()

Parses a given link, and verifies if it’s a valid URL, or a ‘mailto’ link. Returns an iterator of invalid (owner, link) pairs.

class airflow.models.dag.DagTag

Bases: airflow.models.base.Base

A tag name per dag, to allow quick filtering in the DAG view.

__tablename__ = 'dag_tag'

name

dag_id

__repr__()

class airflow.models.dag.DagOwnerAttributes

Bases: airflow.models.base.Base

Table defining different owner attributes. For example, a link for an owner that will be passed as a hyperlink to the DAGs view

__tablename__ = 'dag_owner_attributes'

dag_id

owner

link

__repr__()

classmethod get_all(session)

class airflow.models.dag.DagModel(concurrency=None, **kwargs)

Bases: airflow.models.base.Base

Table containing DAG properties

property next_dagrun_data_interval: airflow.timetables.base.DataInterval | None

property timezone

property safe_dag_id

property relative_fileloc: pathlib.Path | None

File location of the importable dag ‘file’ relative to the configured DAGs folder.

__tablename__ = 'dag'

These items are stored in the database for state related information

dag_id

root_dag_id

is_paused_at_creation

is_paused

is_subdag

is_active

last_parsed_time

last_pickled

last_expired

scheduler_lock

pickle_id

fileloc

processor_subdir

owners

description

default_view

schedule_interval

timetable_description

tags

dag_owner_links

max_active_tasks

max_active_runs

has_task_concurrency_limits

has_import_errors

next_dagrun

next_dagrun_data_interval_start

next_dagrun_data_interval_end

next_dagrun_create_after

__table_args__ = ()

parent_dag

schedule_dataset_references

schedule_datasets

task_outlet_dataset_references

NUM_DAGS_PER_DAGRUN_QUERY

__repr__()

static get_dagmodel(dag_id, session=NEW_SESSION)

classmethod get_current(dag_id, session=NEW_SESSION)

get_last_dagrun(session=NEW_SESSION, include_externally_triggered=False)

get_is_paused(*, session=None)

Provide interface compatibility to ‘DAG’.

static get_paused_dag_ids(dag_ids, session=NEW_SESSION)

Given a list of dag_ids, get a set of Paused Dag Ids

Parameters

• dag_ids (list[str]) – List of Dag ids

• session (sqlalchemy.orm.session.Session) – ORM Session

Returns

Paused Dag_ids

Return type

set[str]

get_default_view()

Get the Default DAG View, returns the default config value if DagModel does not have a value

set_is_paused(is_paused, including_subdags=True, session=NEW_SESSION)

Pause/Un-pause a DAG.

Parameters

• is_paused (bool) – Is the DAG paused

• including_subdags (bool) – whether to include the DAG’s subdags

• session – session

classmethod deactivate_deleted_dags(alive_dag_filelocs, session=NEW_SESSION)

Set is_active=False on the DAGs for which the DAG files have been removed.

Parameters

• alive_dag_filelocs (list[str]) – file paths of alive DAGs

• session – ORM Session

classmethod dags_needing_dagruns(session)

Return (and lock) a list of Dag objects that are due to create a new DagRun.

This will return a resultset of rows that is row-level-locked with a “SELECT … FOR UPDATE” query, you should ensure that any scheduling decisions are made in a single transaction – as soon as the transaction is committed it will be unlocked.

calculate_dagrun_date_fields(dag, most_recent_dag_run)

Calculate next_dagrun and next_dagrun_create_after`

Parameters

• dag (DAG) – The DAG object

• most_recent_dag_run (None | datetime.datetime | airflow.timetables.base.DataInterval) – DataInterval (or datetime) of most recent run of this dag, or none if not yet scheduled.

get_dataset_triggered_next_run_info(*, session=NEW_SESSION)

airflow.models.dag.dag(dag_id='', description=None, schedule=NOTSET, schedule_interval=NOTSET, timetable=None, start_date=None, end_date=None, full_filepath=None, template_searchpath=None, template_undefined=jinja2.StrictUndefined, user_defined_macros=None, user_defined_filters=None, default_args=None, concurrency=None, max_active_tasks=airflow_conf.getint('core', 'max_active_tasks_per_dag'), max_active_runs=airflow_conf.getint('core', 'max_active_runs_per_dag'), dagrun_timeout=None, sla_miss_callback=None, default_view=airflow_conf.get_mandatory_value('webserver', 'dag_default_view').lower(), orientation=airflow_conf.get_mandatory_value('webserver', 'dag_orientation'), catchup=airflow_conf.getboolean('scheduler', 'catchup_by_default'), on_success_callback=None, on_failure_callback=None, doc_md=None, params=None, access_control=None, is_paused_upon_creation=None, jinja_environment_kwargs=None, render_template_as_native_obj=False, tags=None, owner_links=None, auto_register=True)

Python dag decorator. Wraps a function into an Airflow DAG. Accepts kwargs for operator kwarg. Can be used to parameterize DAGs.

Parameters

• dag_args – Arguments for DAG object

• dag_kwargs – Kwargs for DAG object.

class airflow.models.dag.DagContext

DAG context is used to keep the current DAG when DAG is used as ContextManager.

You can use DAG as context:

with DAG(
    dag_id="example_dag",
    default_args=default_args,
    schedule="0 0 * * *",
    dagrun_timeout=timedelta(minutes=60),
) as dag:
    ...

If you do this the context stores the DAG and whenever new task is created, it will use such stored DAG as the parent DAG.

autoregistered_dags: set[tuple[DAG, types.ModuleType]]

current_autoregister_module_name: str | None

classmethod push_context_managed_dag(dag)

classmethod pop_context_managed_dag()

classmethod get_current_dag()

Previous Next

Was this entry helpful?