refactor: unify env-var config in common/config.py

Before: SPARK_EXECUTOR_DATA_DIR, SPARK_EXECUTOR_JOBS_DIR, and
YARN_RESOURCE_MANAGER_URL were each read directly via os.environ.get()
inside the module that used them. Log level was hardcoded DEBUG in
common/logging.py. There was no single file showing what the full set
of env vars the app reads is.

After: common/config.py defines a single Settings dataclass that
reads all env vars at import time and exposes them as fields on a
module-level singleton. App code uses "from common.config import
settings; settings.data_dir" etc. New SPARK_EXECUTOR_LOG_LEVEL env
var controls stderr + info file verbosity (debug file always gets full
DEBUG).

Improvements:
  - One file lists every env var the app reads (was: grep the codebase)
  - Tests can monkeypatch fields on the settings singleton directly
    instead of monkeypatching the env + reloading
  - Adding a new env var means adding one field in config.py, not
    editing 3+ call sites
  - settings.reload() method for tests that prefer env-var style

Out of scope (kept where they are):
  - GUNICORN_* env vars live in gunicorn.conf.py (gunicorn concept)
  - PYTHONUNBUFFERED in Dockerfile (Python runtime flag)
  - SPARK_SUBMIT_OPTS not in config (JVM flag, not Python)

Test changes:
  - test_job_writer.py: settings.jobs_dir instead of monkeypatching
    SPARK_EXECUTOR_JOBS_DIR
  - test_yarn_client.py: settings.yarn_resource_manager_url instead of
    monkeypatching YARN_RESOURCE_MANAGER_URL
  - test_generate_tool.py: same as job_writer
  - Each test file gets an autouse fixture that snapshots+restores
    settings so one test mutation does not leak into the next

116/116 still pass. Live verified: SPARK_EXECUTOR_LOG_LEVEL=INFO
suppresses DEBUG loguru output as expected.
This commit is contained in:
Claude
2026-06-25 10:52:53 +08:00
parent 3ec9ea37fd
commit f8e367d43c
9 changed files with 204 additions and 48 deletions
+10 -3
View File
@@ -6,6 +6,10 @@
Process-wide loguru configuration. Import `logger` from here in every
module instead of instantiating new loggers.
Log level is controlled via `SPARK_EXECUTOR_LOG_LEVEL` (see common/config.py):
DEBUG - default; full verbosity
INFO - quieter; recommended for production
Levels used in this project:
DEBUG - entry/exit of public tools, subprocess commands, file I/O paths
INFO - business events (job submitted, status changed, connection saved)
@@ -15,13 +19,16 @@ Levels used in this project:
import sys
from pathlib import Path
from loguru import logger
from common.config import settings
Path("data/logs/debug").mkdir(parents=True, exist_ok=True)
Path("data/logs/info").mkdir(parents=True, exist_ok=True)
logger.remove()
logger.add(
sys.stderr,
level="DEBUG",
level=settings.log_level,
format=(
"<green>{time:HH:mm:ss.SSS}</green> | "
"<level>{level: <7}</level> | "
@@ -34,7 +41,7 @@ logger.add(
logger.add(
"data/logs/debug/{time:YYYY-MM-DD}.log",
level="DEBUG",
level="DEBUG", # always full at the file level for audit
enqueue=True,
retention="30 days",
compression="gz",
@@ -43,7 +50,7 @@ logger.add(
logger.add(
"data/logs/info/{time:YYYY-MM-DD}.log",
level="INFO",
level=settings.log_level,
enqueue=True,
retention="30 days",
compression="gz",