Develop Auto HDR tests for CI - Basic

[AviaAv]

Tracked on: [LRS-1299]

[Copilot]

Pull Request Overview

This PR implements Auto HDR tests for CI by creating a comprehensive test suite for High Dynamic Range (HDR) functionality. The changes restructure existing HDR tests and add new performance and configuration tests to ensure HDR features work correctly across different scenarios.

• Refactors existing HDR test code into a reusable helper module
• Creates new basic HDR preset tests with manual and auto configurations
• Adds performance testing to validate FPS under HDR operations
• Implements comprehensive configuration testing across multiple resolutions

Reviewed Changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
unit-tests/live/hdr/test-hdr-preset.py	Basic HDR sanity tests for manual and auto modes
unit-tests/live/hdr/test-hdr-performance.py	Performance validation ensuring HDR maintains acceptable FPS
unit-tests/live/hdr/test-hdr-configurations.py	Comprehensive testing across resolutions and HDR disable functionality
unit-tests/live/hdr/hdr_helper.py	Shared helper module with HDR configurations and test utilities
unit-tests/live/d400/test-hdr-preset.py	Removal of old HDR test implementation

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[Copilot]

Using global variables can make code harder to test and maintain. Consider passing batch_size as a parameter or using a class-based approach to encapsulate state.

Suggested change

global batch_size

[Copilot]

Global variable initialization at module level can lead to test interdependencies and state leakage between tests. Consider initializing this within test functions or using a test fixture.

[Copilot]

Using global variables in test functions can cause race conditions and make tests non-deterministic. Consider using a class or passing state through function parameters.

Suggested change

	test_hdr_performance()
	class HDRPerformanceTest:
	EXPECTED_FPS = 30
	ACCEPTABLE_FPS = EXPECTED_FPS * 0.9
	TIME_FOR_STEADY_STATE = 3
	TIME_TO_COUNT_FRAMES = 5
	def __init__(self, sensor):
	self.sensor = sensor
	self.count = 0
	self.count_frames = False
	def frame_counter_callback(self, frame):
	if not self.count_frames:
	return # Skip counting if not enabled
	self.count += 1
	log.d("Frame callback called, frame number:", frame.get_frame_number())
	def run(self):
	"""
	Test HDR performance with various configurations
	"""
	for i, config in enumerate(HDR_CONFIGURATIONS):
	config_type = "Auto" if "depth-ae" in json.dumps(config) else "Manual"
	num_items = len(config["hdr-preset"]["items"])
	test_name = f"Config {i + 1} ({config_type}, {num_items} items)"
	test_json_load(config, test_name)
	self.count = 0
	depth_profile = next(p for p in self.sensor.get_stream_profiles() if p.stream_type() == rs.stream.depth)
	self.sensor.open(depth_profile)
	self.sensor.start(lambda frame: self.frame_counter_callback(frame))
	time.sleep(self.TIME_FOR_STEADY_STATE)
	self.count_frames = True # Start counting frames
	time.sleep(self.TIME_TO_COUNT_FRAMES)
	self.count_frames = False # Stop counting
	self.sensor.stop()
	self.sensor.close()
	measured_fps = self.count / self.TIME_TO_COUNT_FRAMES
	log.d(f"Test {test_name}: Counted frames = {self.count}, Measured FPS = {measured_fps:.2f}")
	test.check(measured_fps > self.ACCEPTABLE_FPS, f"Measured FPS {measured_fps:.2f}")
	hdr_test = HDRPerformanceTest(sensor)
	hdr_test.run()

[Copilot]

The error message should include the expected threshold for clarity. Consider: f"Measured FPS {measured_fps:.2f} is below acceptable threshold {ACCEPTABLE_FPS}"

Suggested change

	test.check(measured_fps > ACCEPTABLE_FPS, f"Measured FPS {measured_fps:.2f}")
	test.check(measured_fps > ACCEPTABLE_FPS, f"Measured FPS {measured_fps:.2f} is below acceptable threshold {ACCEPTABLE_FPS}")

[Copilot]

Global device and context initialization in a helper module can cause issues when multiple tests run concurrently. Consider moving device initialization into individual test functions or using proper test fixtures.