echonet: refactor l1 client, add init, share retry func

[ayeletstarkware]

No description provided.

[reviewable-StarkWare]

This change is 

[ayeletstarkware]

• echonet: parse log to L1Event #10433
• echonet: decode ethereum log into structured event #10431
• echonet: add constants file #10430
• echonet: refactor l1 client, add init, share retry func #10416 👈 (View in Graphite)
• main-v0.14.1

This stack of pull requests is managed by Graphite. Learn more about stacking.

[matanl-starkware]

@matanl-starkware reviewed 1 of 2 files at r1, 1 of 1 files at r2, all commit messages.
Reviewable status: all files reviewed, 4 unresolved discussions (waiting on @ayeletstarkware)

---

echonet/l1_client.py line 41 at r2 (raw file):

        self,
        api_key: str,
        logger: logging.Logger,

Usually, it's the responsibility of the class to create its own logger.
If you want, it's possible to supply a logger "name" externally, but that's also not useful most of the times.

Code quote (i):

logger: logging.Logger,

Code snippet (ii):

self.logger = logging.Logger('L1Client')

---

echonet/l1_client.py line 55 at r2 (raw file):

        self,
        request_func: Callable[[], Dict],
        method_name_for_logs: str,

Consider getting this from the call stack

Suggestion:

import inspect

def caller():
    callee()

def callee():
    caller_frame = inspect.currentframe().f_back
    print(caller_frame.f_code.co_name)

caller()  # prints: caller

---

echonet/l1_client.py line 105 at r2 (raw file):

            response = requests.post(self.rpc_url, json=payload, timeout=self.timeout)
            response.raise_for_status()
            return response.json()

Move the error handling and response parsing to the common retries function.

Suggestion:

request_func = lambda: requests.post(self.rpc_url, json=payload, timeout=self.timeout)

---

echonet/l1_client.py line 106 at r2 (raw file):

            response.raise_for_status()
            return response.json()

Another alternative:
Leave the "timeout" handling to the retries function.

Code snippet:

request_func = functools.partial(requests.post, self.rpc_url, json=payload)

...
response = request_func(timeout=self.timeout)

[ayeletstarkware]

Reviewable status: all files reviewed, 4 unresolved discussions (waiting on @matanl-starkware)

---

echonet/l1_client.py line 41 at r2 (raw file):

Previously, matanl-starkware (Matan Lior) wrote…

Usually, it's the responsibility of the class to create its own logger.
If you want, it's possible to supply a logger "name" externally, but that's also not useful most of the times.

Done.

---

echonet/l1_client.py line 55 at r2 (raw file):

Previously, matanl-starkware (Matan Lior) wrote…

Consider getting this from the call stack

Cool!

---

echonet/l1_client.py line 105 at r2 (raw file):

Previously, matanl-starkware (Matan Lior) wrote…

Move the error handling and response parsing to the common retries function.

used functools

---

echonet/l1_client.py line 106 at r2 (raw file):

Previously, matanl-starkware (Matan Lior) wrote…

Another alternative:
Leave the "timeout" handling to the retries function.

Done.

[matanl-starkware]

@matanl-starkware reviewed 2 of 2 files at r3, all commit messages.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @ayeletstarkware)