Claude
34e6208f54
The agent can now write PySpark that runs DROP/DELETE/UPDATE/etc. on
production tables. Add a static guard that rejects anything other than
SELECT and INSERT at two enforcement points:
1. generate_job_file: validates BEFORE writing to disk. Agent gets
immediate feedback ('rewrite to use only SELECT/INSERT') rather
than learning at submit time.
2. prepare_submit_job: re-validates the script content (reads the
file) as a defense-in-depth check. Catches host-mounted files,
manually-edited files, anything that bypassed generate_job_file.
How it works:
- common/sql_guard.py extracts Python string literals whose first
keyword is a SQL verb (catches spark.sql('...'), f-strings, and any
raw SQL literal)
- sqlparse splits each literal into statements; we check the first
keyword against the policy (SELECT/INSERT/WITH allowed; DROP,
DELETE, UPDATE, TRUNCATE, ALTER, CREATE, REPLACE, MERGE, GRANT,
REVOKE, SET, SHOW, KILL, EXEC, etc. forbidden)
- WITH recurses into the CTE body to catch WITH x AS (DROP ...) ...
- The MCP layer maps ValueError -> HTTP 400 (existing handler)
Test coverage:
- 28 unit tests in test_sql_guard.py cover: extraction (single/double/
f-string, English false positives, multi-literal), statement
classification (SELECT, INSERT, DROP, DELETE, UPDATE, TRUNCATE,
ALTER, CREATE, multi-statement, CTE bodies, comments)
- 2 integration tests verify MCP layer returns 400 with the policy
explanation at both generate_job_file and prepare_submit_job
Limitations (documented in sql_guard.py docstring):
- f-strings where the SQL is built at runtime (e.g. f'SELECT * FROM
{user_input}') look like SELECTs at static-analysis time. The
guard catches the static literal; the runtime substitution is the
caller's responsibility.
- pyspark.sql.functions.expr('...') accepts SQL inline; not currently
caught. (Future work.)
146/146 still pass. Live verified: DROP TABLE -> MCP 400 with policy
explanation; SELECT -> MCP 200 + file written to ./data/jobs/.
44 lines
1.5 KiB
Python
44 lines
1.5 KiB
Python
# coding=utf-8
|
|
"""
|
|
@Time :2026/6/24
|
|
@Author :tao.chen
|
|
"""
|
|
from common.logging import logger
|
|
from common.sql_guard import validate_pyspark_code
|
|
from spark_executor.core.job_writer import write_job_file
|
|
|
|
|
|
class SqlGuardViolation(ValueError):
|
|
"""Raised when generate_job_file receives code that violates the SQL
|
|
safety policy. -> HTTP 400 via the FastAPI ValueError handler.
|
|
"""
|
|
pass
|
|
|
|
|
|
def generate_job_file(code: str) -> dict[str, str]:
|
|
"""Write a PySpark code string to disk; return its absolute path.
|
|
|
|
Validates the code against the SQL safety policy (SELECT/INSERT only)
|
|
BEFORE writing, so the agent gets immediate feedback rather than
|
|
learning at prepare_submit_job time. Use the returned path as the
|
|
`script_path` argument of prepare_submit_job.
|
|
|
|
The output directory is controlled by the SPARK_EXECUTOR_JOBS_DIR env var
|
|
(default: ./data/jobs/).
|
|
"""
|
|
logger.debug(f"generate_job_file enter code_bytes={len(code)}")
|
|
offenses = validate_pyspark_code(code)
|
|
if offenses:
|
|
logger.warning(
|
|
f"generate_job_file rejected: SQL policy violation(s): {offenses}"
|
|
)
|
|
raise SqlGuardViolation(
|
|
f"PySpark code violates SQL safety policy. "
|
|
f"Only SELECT and INSERT statements are allowed. "
|
|
f"Found forbidden statement(s): {offenses}. "
|
|
f"Rewrite the code to use only SELECT/INSERT (or DataFrame DSL)."
|
|
)
|
|
path = write_job_file(code)
|
|
logger.info(f"generate_job_file ok script_path={path}")
|
|
return {"script_path": path}
|