test: add test script #169
Conversation
There was a problem hiding this comment.
When printing use the formatter with print(f'Some thing: {variable_name}')
https://docs.python.org/3/tutorial/inputoutput.html#tut-f-strings
manual-tests/automated-test.py
Outdated
| if ( | ||
| args.zenohd == "/.local/bin/zenohd" | ||
| or args.zfctl == "/.config/zenoh-flow/zfctl.json" | ||
| ): | ||
| print(home_path) | ||
| zf_conf.zenohd_path = home_path + str(args.zenohd) | ||
| zf_conf.zfctl_path = home_path + str(args.zfctl) | ||
| zf_conf.zf_plugin_path = home_path + str(args.plugin) | ||
| else: | ||
| zf_conf.zenohd_path = args.zenohd | ||
| zf_conf.zfctl_path = args.zfctl | ||
| zf_conf.zf_plugin_path = args.plugin |
There was a problem hiding this comment.
It is better to put the correct path in the default options. If users run the "help" option and see "/.config/…" they can assume that something is wrong.
There is the function: os.path.expanduser().
What I suggest you do: provide a default option with ~ in the path & call expanduser in all cases (be it the default option or provided by the user).
manual-tests/automated-test.py
Outdated
| ) | ||
| print(f'show list of Zenoh-Flow instances: \n {zf_command["instances_cmd"]}') | ||
|
|
||
| print("[Info] Looking for Zenohd process...") |
There was a problem hiding this comment.
Given that you are not just looking for zenohd processes, I would rather write:
| print("[Info] Looking for Zenohd process...") | |
| print("[Info] Killing extra `zenohd` processes...") |
manual-tests/automated-test.py
Outdated
| pid = process.pid | ||
| print(f"[Info] kill process: {pid}") | ||
| os.kill(pid, signal.SIGKILL) |
There was a problem hiding this comment.
Couldn't this be simplified like so?
| pid = process.pid | |
| print(f"[Info] kill process: {pid}") | |
| os.kill(pid, signal.SIGKILL) | |
| print(f"[Info] kill process: {process.pid}") | |
| process.kill() |
manual-tests/automated-test.py
Outdated
| print(f'[Info] Launching list instances: {zf_command["instances_cmd"]}') | ||
| launch_instances = os.popen(zf_command["instances_cmd"]) | ||
| count = 0 | ||
| instances_string = launch_instances.read().split() | ||
| for i in range(instances_string.__len__()): | ||
| if instances_string[i][0] == "+": | ||
| count = count + 1 | ||
| if count <= 2: | ||
| print("[Error] test data flow: FAILED.") | ||
| exit() | ||
| else: | ||
| print("[Info] test data flow instance: SUCCESS.") | ||
| instance["uuid_instances"] = instances_string[23] | ||
| instance["flow"] = instances_string[25] | ||
| instance["operators"] = instances_string[27] | ||
| instance["sinks"] = instances_string[29] | ||
| instance["sources"] = instances_string[31] | ||
| instance["connectors"] = instances_string[33] | ||
| instance["links"] = instances_string[35] |
There was a problem hiding this comment.
Remove these lines, they are too specific and will quickly return false negatives.
They are referred to as "magic numbers" and are considered bad practice.
manual-tests/automated-test.py
Outdated
| if str(tuple(data["sources"]).__len__()) == instance["sources"]: | ||
| print( | ||
| f'[Info] test active Source nodes: SUCCESS, the active Source nodes [{str(tuple(data["sources"]).__len__())}] are equal to declared Source nodes [{instance["sources"]}]' | ||
| ) | ||
| if str(tuple(data["sinks"]).__len__()) == instance["sinks"]: | ||
| print( | ||
| f'[Info] test active Sink nodes: SUCCESS, the active Sink nodes [{str(tuple(data["sinks"]).__len__())}] are equal to those declared Sinks nodes [{instance["sinks"]}]' | ||
| ) | ||
| if str(tuple(data["operators"]).__len__()) == instance["operators"]: | ||
| print( | ||
| f'[Info] test active Operator nodes: SUCCESS, the active Operator nodes [{str(tuple(data["operators"]).__len__())}] are equal declared Operator nodes [{instance["operators"]}]' | ||
| ) | ||
| if ( | ||
| str(tuple(data["sources"]).__len__()) != instance["sources"] | ||
| or str(tuple(data["sinks"]).__len__()) != instance["sinks"] | ||
| or str(tuple(data["operators"]).__len__()) != instance["operators"] | ||
| ): | ||
| print("[Error] Test Zenoh-Flow configuration nodesSome nodes: FAILED.") |
There was a problem hiding this comment.
Considering the previous change I asked you to do, these lines need to be removed as well.
manual-tests/automated-test.py
Outdated
| res = txt_res.read() | ||
| if res.split()[1] == "Test!": | ||
| print(f"\n[Info] Test Recive: SUCCESS, {res}") |
There was a problem hiding this comment.
For the test to be more reliable, you need to:
- open the file, count the number of lines,
- do the
curl - ensure that the file contains one more line and check that that extra line is equal to "Hello, Test!"
This pull request wants to add the possibility to test a specific Zenoh-Flow configuration using a script