🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org
Python agentai-societiesartificial-intelligencecommunicative-aicooperative-aideep-learninglarge-language-modelsmulti-agent-systemsnatural-language-processing
# ========= Copyright 2023-2026 @ CAMEL-AI.org. All Rights Reserved. =========
# 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.
# ========= Copyright 2023-2026 @ CAMEL-AI.org. All Rights Reserved. =========
import os
import tempfile
import pytest
from PIL import Image
from camel.tasks import Task, TaskManager
def test_task():
task = Task(content="a start task", id="0")
sub_task_1 = Task(content="a sub task 1", id="1")
sub_task_2 = Task(content="a sub task 2", id="2")
task.add_subtask(sub_task_1)
task.add_subtask(sub_task_2)
task.set_state("DONE")
print(task.to_string())
assert task.state == "DONE"
task.set_state("RUNNING")
print(task.to_string(state=True))
print(task.get_result())
def test_task_management():
"""
Task 0: a start task
Task 1: a sub task 1
Task 2: a sub task 2
Task 2.1: a sub task of task 2
Task 2.2: a sub task of task 2
Task 3: a sub task 3
"""
root_task = Task(content="a start task", id="0")
sub_task_1 = Task(content="a sub task 1", id="1")
sub_task_2 = Task(content="a sub task 2", id="2")
sub_task_3 = Task(content="a sub task 3", id="3")
sub_task_2_1 = Task(content="a sub task of task 2", id="2.1")
sub_task_2_2 = Task(content="a sub task of task 2", id="2.2")
root_task.add_subtask(sub_task_1)
root_task.add_subtask(sub_task_2)
root_task.add_subtask(sub_task_3)
sub_task_2.add_subtask(sub_task_2_1)
sub_task_2.add_subtask(sub_task_2_2)
print("\n===\n" + root_task.to_string() + "\n===\n")
manager = TaskManager(root_task)
new_tasks = [
sub_task_1,
sub_task_2,
sub_task_2_1,
sub_task_2_2,
sub_task_3,
]
sorted_tasks = manager.topological_sort(new_tasks)
sorted_tasks_ids = [task.id for task in sorted_tasks]
print(sorted_tasks_ids)
assert sorted_tasks_ids == ["1", "2.1", "2.2", "2", "3"]
manager.add_tasks(new_tasks)
print([task.id for task in manager.tasks])
assert [task.id for task in manager.tasks] == [
"1",
"2.1",
"2.2",
"2",
"3",
"0",
]
@pytest.fixture
def temp_image():
# Create temporary image file and return PIL Image object
with tempfile.NamedTemporaryFile(suffix=".png", delete=False) as f:
img = Image.new("RGB", (10, 10), color="blue")
img.save(f, format="PNG")
f.flush()
# Load image into memory to avoid file dependency
with Image.open(f.name) as loaded_img:
image = loaded_img.copy()
yield image
os.remove(f.name)
@pytest.fixture
def temp_video_bytes():
# Create temporary video file and read as bytes
with tempfile.NamedTemporaryFile(suffix=".mp4", delete=False) as f:
video_data = b"fake video data"
f.write(video_data)
f.flush()
yield video_data
os.remove(f.name)
def test_image_task_management(temp_image):
"""
Task 0: Analyze image
Task 1: Detect objects in image
Task 2: Recognize text in image
Task 2.1: Recognize title text
Task 2.2: Recognize body text
Task 3: Generate image description
"""
root_task = Task(
content="Analyze image",
id="0",
image_list=[temp_image],
image_detail="high",
)
sub_task_1 = Task(content="Detect objects in image", id="1")
sub_task_2 = Task(content="Recognize text in image", id="2")
sub_task_3 = Task(content="Generate image description", id="3")
sub_task_2_1 = Task(content="Recognize title text", id="2.1")
sub_task_2_2 = Task(content="Recognize body text", id="2.2")
root_task.add_subtask(sub_task_1)
root_task.add_subtask(sub_task_2)
root_task.add_subtask(sub_task_3)
sub_task_2.add_subtask(sub_task_2_1)
sub_task_2.add_subtask(sub_task_2_2)
print("\n===\n" + root_task.to_string() + "\n===\n")
manager = TaskManager(root_task)
new_tasks = [
sub_task_1,
sub_task_2,
sub_task_2_1,
sub_task_2_2,
sub_task_3,
]
sorted_tasks = manager.topological_sort(new_tasks)
sorted_tasks_ids = [task.id for task in sorted_tasks]
print(sorted_tasks_ids)
assert sorted_tasks_ids == ["1", "2.1", "2.2", "2", "3"]
manager.add_tasks(new_tasks)
print([task.id for task in manager.tasks])
assert [task.id for task in manager.tasks] == [
"1",
"2.1",
"2.2",
"2",
"3",
"0",
]
def test_video_task_management(temp_video_bytes):
"""
Task 0: Analyze video
Task 1: Detect people in video
Task 2: Recognize speech in video
Task 2.1: Recognize opening dialogue
Task 2.2: Recognize ending dialogue
Task 3: Generate video summary
"""
root_task = Task(
content="Analyze video",
id="0",
video_bytes=temp_video_bytes,
video_detail="high",
)
sub_task_1 = Task(content="Detect people in video", id="1")
sub_task_2 = Task(content="Recognize speech in video", id="2")
sub_task_3 = Task(content="Generate video summary", id="3")
sub_task_2_1 = Task(content="Recognize opening dialogue", id="2.1")
sub_task_2_2 = Task(content="Recognize ending dialogue", id="2.2")
root_task.add_subtask(sub_task_1)
root_task.add_subtask(sub_task_2)
root_task.add_subtask(sub_task_3)
sub_task_2.add_subtask(sub_task_2_1)
sub_task_2.add_subtask(sub_task_2_2)
print("\n===\n" + root_task.to_string() + "\n===\n")
manager = TaskManager(root_task)
new_tasks = [
sub_task_1,
sub_task_2,
sub_task_2_1,
sub_task_2_2,
sub_task_3,
]
sorted_tasks = manager.topological_sort(new_tasks)
sorted_tasks_ids = [task.id for task in sorted_tasks]
print(sorted_tasks_ids)
assert sorted_tasks_ids == ["1", "2.1", "2.2", "2", "3"]
manager.add_tasks(new_tasks)
print([task.id for task in manager.tasks])
assert [task.id for task in manager.tasks] == [
"1",
"2.1",
"2.2",
"2",
"3",
"0",
]