The Sep 2026 decomposition (PR #102117) makes internal import paths a non-API: names now live in the focused modules that define them. This commit is the ONLY thing keeping the old paths alive, so external plugins have time to update. It is deliberately a single, unsquashed commit: git revert <this sha> removes every shim, stub and manifest at once on the announced date. Nothing in-tree may depend on these pointers: scripts/check_compat_pointers.py (wired into lint.yml) fails CI if it does. What it adds (see COMPAT_MANIFEST.md, compat_manifest.json): - 332 facade modules get one delimited `PLUGIN-COMPAT` block appended at the end of the file - 1,172 moved names resolved lazily via a module `__getattr__` (PEP 562) — never a top-level import, so no import cycles; facades that already had `__getattr__` get a chained one - 592 third-party/stdlib names the old modules used to expose, with their original import statements - 266 public definitions that had been deleted as unused, restored byte-for-byte from the pre-decomposition tree (+40 private helpers and 16 imports pulled in only because a restored definition needs them) - 3 deleted modules recreated as re-export stubs (gateway/startup_watchdog, hermes_cli/observability/ relay_runtime, tools/environments/modal_utils) - private names (`_x`) get no pointer: they were never API (3,792 skipped) Verified: all 335 touched modules import under a fresh HERMES_HOME and every manifest name resolves; the lint reports zero in-tree uses; ruff clean; targeted suites unchanged.
103 lines
3.9 KiB
Python
103 lines
3.9 KiB
Python
"""Bounded reads of HTTP error response bodies.
|
|
|
|
On a non-OK *streaming* response Hermes reads the body for a diagnostic (only ever shown truncated to
|
|
a few hundred chars). A bare ``response.read()`` is unbounded two ways: arbitrarily large body
|
|
(memory) or a body that stalls forever (hang). ``read_streaming_error_body`` caps bytes and enforces a
|
|
hard wall-clock deadline; callers use the returned text instead of ``response.text`` (unbounded /
|
|
raises after a partial stream read). ``httpx.iter_bytes()`` blocks *inside* the socket read, so the
|
|
read runs on a daemon thread; on timeout we close the response (unblocking the read) and return the
|
|
partial bytes. Used by the streaming error-body sites: native Gemini, Gemini Cloud Code, Antigravity.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import logging
|
|
import threading
|
|
from typing import List
|
|
|
|
import httpx
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
# Comfortably holds any real provider error envelope while rejecting pathological bodies.
|
|
DEFAULT_ERROR_BODY_MAX_BYTES = 64 * 1024
|
|
# Hard deadline for the whole read; past it the connection is closed and the partial bytes are kept.
|
|
DEFAULT_ERROR_BODY_TIMEOUT_S = 10.0
|
|
|
|
|
|
def read_streaming_error_body(
|
|
response: httpx.Response,
|
|
*,
|
|
max_bytes: int = DEFAULT_ERROR_BODY_MAX_BYTES,
|
|
timeout_s: float = DEFAULT_ERROR_BODY_TIMEOUT_S,
|
|
) -> str:
|
|
"""Read a non-OK streaming body with a byte cap and a hard deadline.
|
|
|
|
Returns UTF-8 text (errors replaced) truncated to ``max_bytes``. Never raises: transport errors,
|
|
stalls and oversize bodies yield best-effort partial text (or ""), so a read error can't mask the
|
|
original failure.
|
|
"""
|
|
chunks: List[bytes] = []
|
|
state = {"truncated": False}
|
|
done = threading.Event()
|
|
|
|
def _drain() -> None:
|
|
total = 0
|
|
try:
|
|
for chunk in response.iter_bytes():
|
|
if not chunk:
|
|
continue
|
|
remaining = max_bytes - total
|
|
if len(chunk) > remaining:
|
|
if remaining > 0:
|
|
chunks.append(chunk[:remaining])
|
|
state["truncated"] = True
|
|
break
|
|
chunks.append(chunk)
|
|
total += len(chunk)
|
|
except Exception as exc: # noqa: BLE001 - error path must not raise
|
|
logger.debug("bounded error-body read failed: %s", exc)
|
|
finally:
|
|
done.set()
|
|
|
|
threading.Thread(target=_drain, name="bounded-error-body-read", daemon=True).start()
|
|
if not done.wait(timeout=timeout_s):
|
|
logger.debug(
|
|
"bounded error-body read: hard timeout after %.1fs (%d bytes so far)",
|
|
timeout_s, sum(len(c) for c in chunks),
|
|
)
|
|
# Closing cancels any in-flight socket read so the worker unwinds. No join (daemon, may be blocked in C).
|
|
try:
|
|
response.close()
|
|
except Exception: # noqa: BLE001
|
|
pass
|
|
|
|
if state["truncated"]:
|
|
logger.debug(
|
|
"bounded error-body read: capped at %d bytes (max=%d)", sum(len(c) for c in chunks), max_bytes,
|
|
)
|
|
return b"".join(chunks).decode("utf-8", errors="replace")
|
|
|
|
|
|
# ---- BEGIN PLUGIN-COMPAT (revert-scheduled; see COMPAT_MANIFEST.md) ----
|
|
# Names external plugins imported from this module before the Sep 2026 decomposition.
|
|
# Internal code MUST NOT use these (scripts/check_compat_pointers.py fails CI if it does).
|
|
# The whole block is removed by reverting the commit that added it.
|
|
from typing import Optional # noqa: F401,E402
|
|
|
|
def read_error_body_or_default(
|
|
response: httpx.Response,
|
|
*,
|
|
max_bytes: int = DEFAULT_ERROR_BODY_MAX_BYTES,
|
|
timeout_s: float = DEFAULT_ERROR_BODY_TIMEOUT_S,
|
|
) -> Optional[str]:
|
|
"""Like ``read_streaming_error_body`` but returns ``None`` on empty body.
|
|
|
|
Convenience for callers that distinguish "no body" from "empty string".
|
|
"""
|
|
text = read_streaming_error_body(
|
|
response, max_bytes=max_bytes, timeout_s=timeout_s
|
|
)
|
|
return text or None
|
|
# ---- END PLUGIN-COMPAT ----
|