pipelines/sdk/python/kfp/compiler/compiler.py

1051 lines
45 KiB
Python

# Copyright 2021 The Kubeflow Authors
#
# Licensed 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.
"""KFP DSL compiler.
This is an experimental implementation of KFP compiler that compiles KFP
pipeline into Pipeline IR:
https://docs.google.com/document/d/1PUDuSQ8vmeKSBloli53mp7GIvzekaY7sggg6ywy35Dk/
"""
import collections
import inspect
import re
import uuid
from typing import (Any, Callable, Dict, List, Mapping, Optional, Set, Tuple,
Union)
import kfp
from google.protobuf import json_format
from kfp.pipeline_spec import pipeline_spec_pb2
from kfp import dsl
from kfp.compiler import pipeline_spec_builder as builder
from kfp.components import utils as component_utils
from kfp.components import component_factory
from kfp.components import for_loop
from kfp.components import pipeline_context
from kfp.components import pipeline_task
from kfp.components import tasks_group
from kfp.components.types import type_utils
_GroupOrTask = Union[tasks_group.TasksGroup, pipeline_task.PipelineTask]
class Compiler:
"""Experimental DSL compiler that targets the PipelineSpec IR.
It compiles pipeline function into PipelineSpec json string.
PipelineSpec is the IR protobuf message that defines a pipeline:
https://github.com/kubeflow/pipelines/blob/237795539f7b85bac77435e2464367226ee19391/api/v2alpha1/pipeline_spec.proto#L8
In this initial implementation, we only support components authored through
Component yaml spec. And we don't support advanced features like conditions,
static and dynamic loops, etc.
Example::
@dsl.pipeline(
name='name',
description='description',
)
def my_pipeline(a: int = 1, b: str = "default value"):
...
kfp.compiler.Compiler().compile(
pipeline_func=my_pipeline,
package_path='path/to/pipeline.json',
)
"""
def compile(
self,
pipeline_func: Callable[..., Any],
package_path: str,
pipeline_name: Optional[str] = None,
pipeline_parameters: Optional[Mapping[str, Any]] = None,
type_check: bool = True,
) -> None:
"""Compile the given pipeline function into pipeline job json.
Args:
pipeline_func: Pipeline function with @dsl.pipeline decorator.
package_path: The output pipeline spec .json file path. For example,
"~/pipeline_spec.json".
pipeline_name: Optional; the name of the pipeline.
pipeline_parameters: Optional; the mapping from parameter names to
values.
type_check: Optional; whether to enable the type check or not.
Default is True.
"""
type_check_old_value = kfp.TYPE_CHECK
try:
kfp.TYPE_CHECK = type_check
pipeline_spec = self._create_pipeline(
pipeline_func=pipeline_func,
pipeline_name=pipeline_name,
pipeline_parameters_override=pipeline_parameters,
)
self._write_pipeline_spec_json(
pipeline_spec=pipeline_spec,
output_path=package_path,
)
finally:
kfp.TYPE_CHECK = type_check_old_value
def _create_pipeline(
self,
pipeline_func: Callable[..., Any],
pipeline_name: Optional[str] = None,
pipeline_parameters_override: Optional[Mapping[str, Any]] = None,
) -> pipeline_spec_pb2.PipelineSpec:
"""Creates a pipeline instance and constructs the pipeline spec from
it.
Args:
pipeline_func: The pipeline function with @dsl.pipeline decorator.
pipeline_name: Optional; the name of the pipeline.
pipeline_parameters_override: Optional; the mapping from parameter
names to values.
Returns:
A PipelineSpec proto representing the compiled pipeline.
"""
# Create the arg list with no default values and call pipeline function.
# Assign type information to the PipelineChannel
pipeline_meta = component_factory.extract_component_interface(
pipeline_func)
pipeline_name = pipeline_name or pipeline_meta.name
pipeline_root = getattr(pipeline_func, 'pipeline_root', None)
args_list = []
signature = inspect.signature(pipeline_func)
for arg_name in signature.parameters:
arg_type = pipeline_meta.inputs[arg_name].type
if not type_utils.is_parameter_type(arg_type):
raise TypeError(
'The pipeline argument "{arg_name}" is viewed as an artifact'
' due to its type "{arg_type}". And we currently do not '
'support passing artifacts as pipeline inputs. Consider type'
' annotating the argument with a primitive type, such as '
'"str", "int", "float", "bool", "dict", and "list".'.format(
arg_name=arg_name, arg_type=arg_type))
args_list.append(
dsl.PipelineParameterChannel(
name=arg_name, channel_type=arg_type))
with pipeline_context.Pipeline(pipeline_name) as dsl_pipeline:
pipeline_func(*args_list)
if not dsl_pipeline.tasks:
raise ValueError('Task is missing from pipeline.')
self._validate_exit_handler(dsl_pipeline)
pipeline_inputs = pipeline_meta.inputs or {}
# Verify that pipeline_parameters_override contains only input names
# that match the pipeline inputs definition.
pipeline_parameters_override = pipeline_parameters_override or {}
for input_name in pipeline_parameters_override:
if input_name not in pipeline_inputs:
raise ValueError(
'Pipeline parameter {} does not match any known '
'pipeline argument.'.format(input_name))
# Fill in the default values.
args_list_with_defaults = [
dsl.PipelineParameterChannel(
name=input_name,
channel_type=input_spec.type,
value=pipeline_parameters_override.get(input_name) or
input_spec.default,
) for input_name, input_spec in pipeline_inputs.items()
]
# Making the pipeline group name unique to prevent name clashes with
# templates
pipeline_group = dsl_pipeline.groups[0]
pipeline_group.name = uuid.uuid4().hex
pipeline_spec = self._create_pipeline_spec(
pipeline_args=args_list_with_defaults,
pipeline=dsl_pipeline,
)
if pipeline_root:
pipeline_spec.default_pipeline_root = pipeline_root
return pipeline_spec
def _write_pipeline_spec_json(
self,
pipeline_spec: pipeline_spec_pb2.PipelineSpec,
output_path: str,
) -> None:
"""Writes pipeline spec into a json file.
Args:
pipeline_spec: IR pipeline spec.
ouput_path: The file path to be written.
Raises:
ValueError: if the specified output path doesn't end with the
acceptable extention.
"""
json_text = json_format.MessageToJson(pipeline_spec, sort_keys=True)
if output_path.endswith('.json'):
with open(output_path, 'w') as json_file:
json_file.write(json_text)
else:
raise ValueError(
'The output path {} should ends with ".json".'.format(
output_path))
def _validate_exit_handler(self,
pipeline: pipeline_context.Pipeline) -> None:
"""Makes sure there is only one global exit handler.
This is temporary to be compatible with KFP v1.
Raises:
ValueError if there are more than one exit handler.
"""
def _validate_exit_handler_helper(
group: tasks_group.TasksGroup,
exiting_task_names: List[str],
handler_exists: bool,
) -> None:
if isinstance(group, dsl.ExitHandler):
if handler_exists or len(exiting_task_names) > 1:
raise ValueError(
'Only one global exit_handler is allowed and all ops need to be included.'
)
handler_exists = True
if group.tasks:
exiting_task_names.extend([x.name for x in group.tasks])
for group in group.groups:
_validate_exit_handler_helper(
group=group,
exiting_task_names=exiting_task_names,
handler_exists=handler_exists,
)
_validate_exit_handler_helper(
group=pipeline.groups[0],
exiting_task_names=[],
handler_exists=False,
)
def _create_pipeline_spec(
self,
pipeline_args: List[dsl.PipelineChannel],
pipeline: pipeline_context.Pipeline,
) -> pipeline_spec_pb2.PipelineSpec:
"""Creates a pipeline spec object.
Args:
pipeline_args: The list of pipeline input parameters.
pipeline: The instantiated pipeline object.
Returns:
A PipelineSpec proto representing the compiled pipeline.
Raises:
ValueError if the argument is of unsupported types.
"""
self._validate_pipeline_name(pipeline.name)
deployment_config = pipeline_spec_pb2.PipelineDeploymentConfig()
pipeline_spec = pipeline_spec_pb2.PipelineSpec()
pipeline_spec.pipeline_info.name = pipeline.name
pipeline_spec.sdk_version = 'kfp-{}'.format(kfp.__version__)
# Schema version 2.1.0 is required for kfp-pipeline-spec>0.1.13
pipeline_spec.schema_version = '2.1.0'
pipeline_spec.root.CopyFrom(
builder.build_component_spec_for_group(
pipeline_channels=pipeline_args,
is_root_group=True,
))
root_group = pipeline.groups[0]
all_groups = self._get_all_groups(root_group)
group_name_to_group = {group.name: group for group in all_groups}
task_name_to_parent_groups, group_name_to_parent_groups = (
self._get_parent_groups(root_group))
condition_channels = self._get_condition_channels_for_tasks(root_group)
name_to_for_loop_group = {
group_name: group
for group_name, group in group_name_to_group.items()
if isinstance(group, dsl.ParallelFor)
}
inputs = self._get_inputs_for_all_groups(
pipeline=pipeline,
pipeline_args=pipeline_args,
root_group=root_group,
task_name_to_parent_groups=task_name_to_parent_groups,
group_name_to_parent_groups=group_name_to_parent_groups,
condition_channels=condition_channels,
name_to_for_loop_group=name_to_for_loop_group,
)
dependencies = self._get_dependencies(
pipeline=pipeline,
root_group=root_group,
task_name_to_parent_groups=task_name_to_parent_groups,
group_name_to_parent_groups=group_name_to_parent_groups,
group_name_to_group=group_name_to_group,
condition_channels=condition_channels,
)
for group in all_groups:
self._build_spec_by_group(
pipeline_spec=pipeline_spec,
deployment_config=deployment_config,
group=group,
inputs=inputs,
dependencies=dependencies,
rootgroup_name=root_group.name,
task_name_to_parent_groups=task_name_to_parent_groups,
group_name_to_parent_groups=group_name_to_parent_groups,
name_to_for_loop_group=name_to_for_loop_group,
)
# TODO: refactor to support multiple exit handler per pipeline.
if pipeline.groups[0].groups:
first_group = pipeline.groups[0].groups[0]
if isinstance(first_group, dsl.ExitHandler):
exit_task = first_group.exit_task
exit_task_name = component_utils.sanitize_task_name(
exit_task.name)
exit_handler_group_task_name = component_utils.sanitize_task_name(
first_group.name)
input_parameters_in_current_dag = [
input_name for input_name in
pipeline_spec.root.input_definitions.parameters
]
exit_task_task_spec = builder.build_task_spec_for_exit_task(
task=exit_task,
dependent_task=exit_handler_group_task_name,
pipeline_inputs=pipeline_spec.root.input_definitions,
)
exit_task_component_spec = builder.build_component_spec_for_exit_task(
task=exit_task)
exit_task_container_spec = builder.build_container_spec_for_task(
task=exit_task)
# Add exit task task spec
pipeline_spec.root.dag.tasks[exit_task_name].CopyFrom(
exit_task_task_spec)
# Add exit task component spec if it does not exist.
component_name = exit_task_task_spec.component_ref.name
if component_name not in pipeline_spec.components:
pipeline_spec.components[component_name].CopyFrom(
exit_task_component_spec)
# Add exit task container spec if it does not exist.
executor_label = exit_task_component_spec.executor_label
if executor_label not in deployment_config.executors:
deployment_config.executors[
executor_label].container.CopyFrom(
exit_task_container_spec)
pipeline_spec.deployment_spec.update(
json_format.MessageToDict(deployment_config))
return pipeline_spec
def _validate_pipeline_name(self, name: str) -> None:
"""Validate pipeline name.
A valid pipeline name should match ^[a-z0-9][a-z0-9-]{0,127}$.
Args:
name: The pipeline name.
Raises:
ValueError if the pipeline name doesn't conform to the regular expression.
"""
pattern = re.compile(r'^[a-z0-9][a-z0-9-]{0,127}$')
if not pattern.match(name):
raise ValueError(
'Invalid pipeline name: %s.\n'
'Please specify a pipeline name that matches the regular '
'expression "^[a-z0-9][a-z0-9-]{0,127}$" using '
'`dsl.pipeline(name=...)` decorator.' % name)
def _get_all_groups(
self,
root_group: tasks_group.TasksGroup,
) -> List[tasks_group.TasksGroup]:
"""Gets all groups (not including tasks) in a pipeline.
Args:
root_group: The root group of a pipeline.
Returns:
A list of all groups in topological order (parent first).
"""
all_groups = []
def _get_all_groups_helper(
group: tasks_group.TasksGroup,
all_groups: List[tasks_group.TasksGroup],
):
all_groups.append(group)
for group in group.groups:
_get_all_groups_helper(group, all_groups)
_get_all_groups_helper(root_group, all_groups)
return all_groups
def _get_parent_groups(
self,
root_group: tasks_group.TasksGroup,
) -> Tuple[Mapping[str, List[_GroupOrTask]], Mapping[str,
List[_GroupOrTask]]]:
"""Get parent groups that contain the specified tasks.
Each pipeline has a root group. Each group has a list of tasks (leaf)
and groups.
This function traverse the tree and get ancestor groups for all tasks.
Args:
root_group: The root group of a pipeline.
Returns:
A tuple. The first item is a mapping of task names to parent groups,
and second item is a mapping of group names to parent groups.
A list of parent groups is a list of ancestor groups including the
task/group itself. The list is sorted in a way that the farthest
parent group is the first and task/group itself is the last.
"""
def _get_parent_groups_helper(
current_groups: List[tasks_group.TasksGroup],
tasks_to_groups: Dict[str, List[_GroupOrTask]],
groups_to_groups: Dict[str, List[_GroupOrTask]],
) -> None:
root_group = current_groups[-1]
for group in root_group.groups:
groups_to_groups[group.name] = [x.name for x in current_groups
] + [group.name]
current_groups.append(group)
_get_parent_groups_helper(
current_groups=current_groups,
tasks_to_groups=tasks_to_groups,
groups_to_groups=groups_to_groups,
)
del current_groups[-1]
for task in root_group.tasks:
tasks_to_groups[task.name] = [x.name for x in current_groups
] + [task.name]
tasks_to_groups = {}
groups_to_groups = {}
current_groups = [root_group]
_get_parent_groups_helper(
current_groups=current_groups,
tasks_to_groups=tasks_to_groups,
groups_to_groups=groups_to_groups,
)
return (tasks_to_groups, groups_to_groups)
# TODO: do we really need this?
def _get_condition_channels_for_tasks(
self,
root_group: tasks_group.TasksGroup,
) -> Mapping[str, Set[dsl.PipelineChannel]]:
"""Gets channels referenced in conditions of tasks' parents.
Args:
root_group: The root group of a pipeline.
Returns:
A mapping of task name to a set of pipeline channels appeared in its
parent dsl.Condition groups.
"""
conditions = collections.defaultdict(set)
def _get_condition_channels_for_tasks_helper(
group,
current_conditions_channels,
):
new_current_conditions_channels = current_conditions_channels
if isinstance(group, dsl.Condition):
new_current_conditions_channels = list(
current_conditions_channels)
if isinstance(group.condition.left_operand,
dsl.PipelineChannel):
new_current_conditions_channels.append(
group.condition.left_operand)
if isinstance(group.condition.right_operand,
dsl.PipelineChannel):
new_current_conditions_channels.append(
group.condition.right_operand)
for task in group.tasks:
for channel in new_current_conditions_channels:
conditions[task.name].add(channel)
for group in group.groups:
_get_condition_channels_for_tasks_helper(
group, new_current_conditions_channels)
_get_condition_channels_for_tasks_helper(root_group, [])
return conditions
def _get_inputs_for_all_groups(
self,
pipeline: pipeline_context.Pipeline,
pipeline_args: List[dsl.PipelineChannel],
root_group: tasks_group.TasksGroup,
task_name_to_parent_groups: Mapping[str, List[_GroupOrTask]],
group_name_to_parent_groups: Mapping[str, List[tasks_group.TasksGroup]],
condition_channels: Mapping[str, Set[dsl.PipelineParameterChannel]],
name_to_for_loop_group: Mapping[str, dsl.ParallelFor],
) -> Mapping[str, List[Tuple[dsl.PipelineChannel, str]]]:
"""Get inputs and outputs of each group and op.
Args:
pipeline: The instantiated pipeline object.
pipeline_args: The list of pipeline function arguments as
PipelineChannel.
root_group: The root group of the pipeline.
task_name_to_parent_groups: The dict of task name to list of parent
groups.
group_name_to_parent_groups: The dict of group name to list of
parent groups.
condition_channels: The dict of task name to a set of pipeline
channels referenced by its parent condition groups.
name_to_for_loop_group: The dict of for loop group name to loop
group.
Returns:
A mapping with key being the group/task names and values being list
of tuples (channel, producing_task_name).
producing_task_name is the name of the task that produces the
channel. If the channel is a pipeline argument (no producer task),
then producing_task_name is None.
"""
inputs = collections.defaultdict(set)
for task in pipeline.tasks.values():
# task's inputs and all channels used in conditions for that task are
# considered.
task_condition_inputs = list(condition_channels[task.name])
for channel in task.channel_inputs + task_condition_inputs:
# If the value is already provided (immediate value), then no
# need to expose it as input for its parent groups.
if getattr(channel, 'value', None):
continue
# channels_to_add could be a list of PipelineChannels when loop
# args are involved. Given a nested loops example as follows:
#
# def my_pipeline(loop_parameter: list):
# with dsl.ParallelFor(loop_parameter) as item:
# with dsl.ParallelFor(item.p_a) as item_p_a:
# print_op(item_p_a.q_a)
#
# The print_op takes an input of
# {{channel:task=;name=loop_parameter-loop-item-subvar-p_a-loop-item-subvar-q_a;}}.
# Given this, we calculate the list of PipelineChannels potentially
# needed by across DAG levels as follows:
#
# [{{channel:task=;name=loop_parameter-loop-item-subvar-p_a-loop-item-subvar-q_a}},
# {{channel:task=;name=loop_parameter-loop-item-subvar-p_a-loop-item}},
# {{channel:task=;name=loop_parameter-loop-item-subvar-p_a}},
# {{channel:task=;name=loop_parameter-loop-item}},
# {{chaenel:task=;name=loop_parameter}}]
#
# For the above example, the first loop needs the input of
# {{channel:task=;name=loop_parameter}},
# the second loop needs the input of
# {{channel:task=;name=loop_parameter-loop-item}}
# and the print_op needs the input of
# {{channel:task=;name=loop_parameter-loop-item-subvar-p_a-loop-item}}
#
# When we traverse a DAG in a top-down direction, we add channels
# from the end, and pop it out when it's no longer needed by the
# sub-DAG.
# When we traverse a DAG in a bottom-up direction, we add
# channels from the front, and pop it out when it's no longer
# needed by the parent DAG.
channels_to_add = collections.deque()
channel_to_add = channel
while isinstance(channel_to_add, (
for_loop.LoopArgument,
for_loop.LoopArgumentVariable,
)):
channels_to_add.append(channel_to_add)
if isinstance(channel_to_add,
for_loop.LoopArgumentVariable):
channel_to_add = channel_to_add.loop_argument
else:
channel_to_add = channel_to_add.items_or_pipeline_channel
if isinstance(channel_to_add, dsl.PipelineChannel):
channels_to_add.append(channel_to_add)
if channel.task_name:
# The PipelineChannel is produced by a task.
upstream_task = pipeline.tasks[channel.task_name]
upstream_groups, downstream_groups = (
self._get_uncommon_ancestors(
task_name_to_parent_groups=task_name_to_parent_groups,
group_name_to_parent_groups=group_name_to_parent_groups,
task1=upstream_task,
task2=task,
))
for i, group_name in enumerate(downstream_groups):
if i == 0:
# If it is the first uncommon downstream group, then
# the input comes from the first uncommon upstream
# group.
producer_task = upstream_groups[0]
else:
# If not the first downstream group, then the input
# is passed down from its ancestor groups so the
# upstream group is None.
producer_task = None
inputs[group_name].add(
(channels_to_add[-1], producer_task))
if group_name in name_to_for_loop_group:
loop_group = name_to_for_loop_group[group_name]
# Pop out the last elements from channels_to_add if it
# is found in the current (loop) DAG. Downstreams
# would only need the more specific versions for it.
if channels_to_add[
-1].full_name in loop_group.loop_argument.full_name:
channels_to_add.pop()
if not channels_to_add:
break
else:
# The PipelineChannel is not produced by a task. It's either
# a top-level pipeline input, or a constant value to loop
# items.
# TODO: revisit if this is correct.
if getattr(task, 'is_exit_handler', False):
continue
# For PipelineChannel as a result of constant value used as
# loop items, we have to go from bottom-up because the
# PipelineChannel can be originated from the middle a DAG,
# which is not needed and visible to its parent DAG.
if isinstance(
channel,
(for_loop.LoopArgument, for_loop.LoopArgumentVariable
)) and channel.is_with_items_loop_argument:
for group_name in task_name_to_parent_groups[
task.name][::-1]:
inputs[group_name].add((channels_to_add[0], None))
if group_name in name_to_for_loop_group:
# for example:
# loop_group.loop_argument.name = 'loop-item-param-1'
# channel.name = 'loop-item-param-1-subvar-a'
loop_group = name_to_for_loop_group[group_name]
if channels_to_add[
0].full_name in loop_group.loop_argument.full_name:
channels_to_add.popleft()
if not channels_to_add:
break
else:
# For PipelineChannel from pipeline input, go top-down
# just like we do for PipelineChannel produced by a task.
for group_name in task_name_to_parent_groups[task.name]:
inputs[group_name].add((channels_to_add[-1], None))
if group_name in name_to_for_loop_group:
loop_group = name_to_for_loop_group[group_name]
if channels_to_add[
-1].full_name in loop_group.loop_argument.full_name:
channels_to_add.pop()
if not channels_to_add:
break
return inputs
def _get_uncommon_ancestors(
self,
task_name_to_parent_groups: Mapping[str, List[_GroupOrTask]],
group_name_to_parent_groups: Mapping[str, List[tasks_group.TasksGroup]],
task1: _GroupOrTask,
task2: _GroupOrTask,
) -> Tuple[List[_GroupOrTask], List[_GroupOrTask]]:
"""Gets the unique ancestors between two tasks.
For example, task1's ancestor groups are [root, G1, G2, G3, task1],
task2's ancestor groups are [root, G1, G4, task2], then it returns a
tuple ([G2, G3, task1], [G4, task2]).
Args:
task_name_to_parent_groups: The dict of task name to list of parent
groups.
group_name_tor_parent_groups: The dict of group name to list of
parent groups.
task1: One of the two tasks.
task2: The other task.
Returns:
A tuple which are lists of uncommon ancestors for each task.
"""
if task1.name in task_name_to_parent_groups:
task1_groups = task_name_to_parent_groups[task1.name]
elif task1.name in group_name_to_parent_groups:
task1_groups = group_name_to_parent_groups[task1.name]
else:
raise ValueError(task1.name + ' does not exist.')
if task2.name in task_name_to_parent_groups:
task2_groups = task_name_to_parent_groups[task2.name]
elif task2.name in group_name_to_parent_groups:
task2_groups = group_name_to_parent_groups[task2.name]
else:
raise ValueError(task2.name + ' does not exist.')
both_groups = [task1_groups, task2_groups]
common_groups_len = sum(
1 for x in zip(*both_groups) if x == (x[0],) * len(x))
group1 = task1_groups[common_groups_len:]
group2 = task2_groups[common_groups_len:]
return (group1, group2)
def _get_dependencies(
self,
pipeline: pipeline_context.Pipeline,
root_group: tasks_group.TasksGroup,
task_name_to_parent_groups: Mapping[str, List[_GroupOrTask]],
group_name_to_parent_groups: Mapping[str, List[tasks_group.TasksGroup]],
group_name_to_group: Mapping[str, tasks_group.TasksGroup],
condition_channels: Dict[str, dsl.PipelineChannel],
) -> Mapping[str, List[_GroupOrTask]]:
"""Gets dependent groups and tasks for all tasks and groups.
Args:
pipeline: The instantiated pipeline object.
root_group: The root group of the pipeline.
task_name_to_parent_groups: The dict of task name to list of parent
groups.
group_name_to_parent_groups: The dict of group name to list of
parent groups.
group_name_to_group: The dict of group name to group.
condition_channels: The dict of task name to a set of pipeline
channels referenced by its parent condition groups.
Returns:
A Mapping where key is group/task name, value is a list of dependent
groups/tasks. The dependencies are calculated in the following way:
if task2 depends on task1, and their ancestors are
[root, G1, G2, task1] and [root, G1, G3, G4, task2], then G3 is
dependent on G2. Basically dependency only exists in the first
uncommon ancesters in their ancesters chain. Only sibling
groups/tasks can have dependencies.
Raises:
RuntimeError: if a task depends on a task inside a condition or loop
group.
"""
dependencies = collections.defaultdict(set)
for task in pipeline.tasks.values():
upstream_task_names = set()
task_condition_inputs = list(condition_channels[task.name])
for channel in task.channel_inputs + task_condition_inputs:
if channel.task_name:
upstream_task_names.add(channel.task_name)
upstream_task_names |= set(task.dependent_tasks)
for upstream_task_name in upstream_task_names:
# the dependent op could be either a BaseOp or an opsgroup
if upstream_task_name in pipeline.tasks:
upstream_task = pipeline.tasks[upstream_task_name]
elif upstream_task_name in group_name_to_group:
upstream_task = group_name_to_group[upstream_task_name]
else:
raise ValueError(
f'Compiler cannot find task: {upstream_task_name}.')
upstream_groups, downstream_groups = self._get_uncommon_ancestors(
task_name_to_parent_groups=task_name_to_parent_groups,
group_name_to_parent_groups=group_name_to_parent_groups,
task1=upstream_task,
task2=task,
)
# If a task depends on a condition group or a loop group, it
# must explicitly dependent on a task inside the group. This
# should not be allowed, because it leads to ambiguous
# expectations for runtime behaviors.
dependent_group = group_name_to_group.get(
upstream_groups[0], None)
if isinstance(dependent_group,
(tasks_group.Condition, tasks_group.ParallelFor)):
raise RuntimeError(
f'Task {task.name} cannot dependent on any task inside'
f' the group: {upstream_groups[0]}.')
dependencies[downstream_groups[0]].add(upstream_groups[0])
return dependencies
def _build_spec_by_group(
self,
pipeline_spec: pipeline_spec_pb2.PipelineSpec,
deployment_config: pipeline_spec_pb2.PipelineDeploymentConfig,
group: tasks_group.TasksGroup,
inputs: Mapping[str, List[Tuple[dsl.PipelineChannel, str]]],
dependencies: Dict[str, List[_GroupOrTask]],
rootgroup_name: str,
task_name_to_parent_groups: Mapping[str, List[_GroupOrTask]],
group_name_to_parent_groups: Mapping[str, List[tasks_group.TasksGroup]],
name_to_for_loop_group: Mapping[str, dsl.ParallelFor],
) -> None:
"""Generates IR spec given a TasksGroup.
Args:
pipeline_spec: The pipeline_spec to update in place.
deployment_config: The deployment_config to hold all executors. The
spec is updated in place.
group: The TasksGroup to generate spec for.
inputs: The inputs dictionary. The keys are group/task names and the
values are lists of tuples (channel, producing_task_name).
dependencies: The group dependencies dictionary. The keys are group
or task names, and the values are lists of dependent groups or
tasks.
rootgroup_name: The name of the group root. Used to determine whether
the component spec for the current group should be the root dag.
task_name_to_parent_groups: The dict of task name to parent groups.
Key is task name. Value is a list of ancestor groups including
the task itself. The list of a given task is sorted in a way that
the farthest group is the first and the task itself is the last.
group_name_to_parent_groups: The dict of group name to parent groups.
Key is the group name. Value is a list of ancestor groups
including the group itself. The list of a given group is sorted
in a way that the farthest group is the first and the group
itself is the last.
name_to_for_loop_group: The dict of for loop group name to loop
group.
"""
group_component_name = component_utils.sanitize_component_name(
group.name)
if group.name == rootgroup_name:
group_component_spec = pipeline_spec.root
else:
group_component_spec = pipeline_spec.components[
group_component_name]
task_name_to_task_spec = {}
task_name_to_component_spec = {}
# Generate task specs and component specs for the dag.
subgroups = group.groups + group.tasks
for subgroup in subgroups:
subgroup_inputs = inputs.get(subgroup.name, [])
subgroup_channels = [channel for channel, _ in subgroup_inputs]
subgroup_component_name = (
component_utils.sanitize_component_name(subgroup.name))
tasks_in_current_dag = [
component_utils.sanitize_task_name(subgroup.name)
for subgroup in subgroups
]
input_parameters_in_current_dag = [
input_name for input_name in
group_component_spec.input_definitions.parameters
]
input_artifacts_in_current_dag = [
input_name for input_name in
group_component_spec.input_definitions.artifacts
]
is_parent_component_root = (
group_component_spec == pipeline_spec.root)
if isinstance(subgroup, pipeline_task.PipelineTask):
subgroup_task_spec = builder.build_task_spec_for_task(
task=subgroup,
parent_component_inputs=group_component_spec
.input_definitions,
tasks_in_current_dag=tasks_in_current_dag,
input_parameters_in_current_dag=input_parameters_in_current_dag,
input_artifacts_in_current_dag=input_artifacts_in_current_dag,
)
task_name_to_task_spec[subgroup.name] = subgroup_task_spec
subgroup_component_spec = builder.build_component_spec_for_task(
task=subgroup)
task_name_to_component_spec[
subgroup.name] = subgroup_component_spec
executor_label = subgroup_component_spec.executor_label
if executor_label not in deployment_config.executors:
if subgroup.container_spec is not None:
subgroup_container_spec = builder.build_container_spec_for_task(
task=subgroup)
deployment_config.executors[
executor_label].container.CopyFrom(
subgroup_container_spec)
elif subgroup.importer_spec is not None:
subgroup_importer_spec = builder.build_importer_spec_for_task(
task=subgroup)
deployment_config.executors[
executor_label].importer.CopyFrom(
subgroup_importer_spec)
elif isinstance(subgroup, dsl.ParallelFor):
# "Punch the hole", adding additional inputs (other than loop
# arguments which will be handled separately) needed by its
# subgroups or tasks.
loop_subgroup_channels = []
for channel in subgroup_channels:
# Skip 'withItems' loop arguments if it's from an inner loop.
if isinstance(
channel,
(for_loop.LoopArgument, for_loop.LoopArgumentVariable
)) and channel.is_with_items_loop_argument:
withitems_loop_arg_found_in_self_or_upstream = False
for group_name in group_name_to_parent_groups[
subgroup.name][::-1]:
if group_name in name_to_for_loop_group:
loop_group = name_to_for_loop_group[group_name]
if channel.name in loop_group.loop_argument.name:
withitems_loop_arg_found_in_self_or_upstream = True
break
if not withitems_loop_arg_found_in_self_or_upstream:
continue
loop_subgroup_channels.append(channel)
if subgroup.items_is_pipeline_channel:
# This loop_argument is based on a pipeline channel, i.e.,
# rather than a static list, it is either the output of
# another task or an input as global pipeline parameters.
loop_subgroup_channels.append(
subgroup.loop_argument.items_or_pipeline_channel)
loop_subgroup_channels.append(subgroup.loop_argument)
subgroup_component_spec = builder.build_component_spec_for_group(
pipeline_channels=loop_subgroup_channels,
is_root_group=False,
)
subgroup_task_spec = builder.build_task_spec_for_group(
group=subgroup,
pipeline_channels=loop_subgroup_channels,
tasks_in_current_dag=tasks_in_current_dag,
is_parent_component_root=is_parent_component_root,
)
elif isinstance(subgroup, dsl.Condition):
# "Punch the hole", adding inputs needed by its subgroups or
# tasks.
condition_subgroup_channels = list(subgroup_channels)
for operand in [
subgroup.condition.left_operand,
subgroup.condition.right_operand,
]:
if isinstance(operand, dsl.PipelineChannel):
condition_subgroup_channels.append(operand)
subgroup_component_spec = builder.build_component_spec_for_group(
pipeline_channels=condition_subgroup_channels,
is_root_group=False,
)
subgroup_task_spec = builder.build_task_spec_for_group(
group=subgroup,
pipeline_channels=condition_subgroup_channels,
tasks_in_current_dag=tasks_in_current_dag,
is_parent_component_root=is_parent_component_root,
)
elif isinstance(subgroup, dsl.ExitHandler):
subgroup_component_spec = builder.build_component_spec_for_group(
pipeline_channels=subgroup_channels,
is_root_group=False,
)
subgroup_task_spec = builder.build_task_spec_for_group(
group=subgroup,
pipeline_channels=subgroup_channels,
tasks_in_current_dag=tasks_in_current_dag,
is_parent_component_root=is_parent_component_root,
)
else:
raise RuntimeError(
f'Unexpected task/group type: Got {subgroup} of type '
f'{type(subgroup)}.')
# Generate dependencies section for this task.
if dependencies.get(subgroup.name, None):
group_dependencies = list(dependencies[subgroup.name])
group_dependencies.sort()
subgroup_task_spec.dependent_tasks.extend([
component_utils.sanitize_task_name(dep)
for dep in group_dependencies
])
# Add component spec if not exists
if subgroup_component_name not in pipeline_spec.components:
pipeline_spec.components[subgroup_component_name].CopyFrom(
subgroup_component_spec)
# Add task spec
group_component_spec.dag.tasks[subgroup.name].CopyFrom(
subgroup_task_spec)
pipeline_spec.deployment_spec.update(
json_format.MessageToDict(deployment_config))
# Surface metrics outputs to the top.
builder.populate_metrics_in_dag_outputs(
tasks=group.tasks,
task_name_to_parent_groups=task_name_to_parent_groups,
task_name_to_task_spec=task_name_to_task_spec,
task_name_to_component_spec=task_name_to_component_spec,
pipeline_spec=pipeline_spec,
)