feat(windows): close remaining POSIX-only landmines — TUI crash, kanban waitpid, AF_UNIX sandbox, /bin/bash, npm .cmd shims, cwd tracking, detach flags

Second pass on native Windows support, driven by a systematic audit across
five areas: POSIX-only primitives (signal.SIGKILL/SIGHUP/SIGPIPE, os.WNOHANG,
os.setsid), path translation bugs (/c/Users → C:\Users), subprocess patterns
(npm.cmd batch shims, start_new_session no-op on Windows), subsystem health
(cron, gateway daemon, update flow), and module-level import guards.

Every change is platform-gated — POSIX (Linux/macOS) behaviour is preserved
bit-identical. Explicit "do no harm" test: test_posix_path_preserved_on_linux,
test_posix_noop, test_windows_detach_popen_kwargs_is_posix_equivalent_on_posix.

## New module

- hermes_cli/_subprocess_compat.py — shared helpers (resolve_node_command,
  windows_detach_flags, windows_hide_flags, windows_detach_popen_kwargs).
  All no-ops on non-Windows.

## CRITICAL fixes (would crash or silently break on Windows)

- tui_gateway/entry.py: SIGPIPE/SIGHUP referenced at module top level would
  AttributeError on import on Windows, breaking `hermes --tui` entirely (it
  spawns this module as a subprocess).  Guard each signal.signal() call with
  hasattr() and add SIGBREAK as Windows' SIGHUP equivalent.

- hermes_cli/kanban_db.py: os.waitpid(-1, os.WNOHANG) in dispatcher tick was
  unguarded.  os.WNOHANG doesn't exist on Windows.  Gate the whole reap loop
  behind `os.name != "nt"` — Windows has no zombies anyway.

- tools/code_execution_tool.py: AF_UNIX socket for execute_code RPC fails on
  most Windows builds.  Fall back to loopback TCP (AF_INET on 127.0.0.1:0
  ephemeral port) when _IS_WINDOWS.  HERMES_RPC_SOCKET env var now accepts
  either a filesystem path (POSIX) or `tcp://127.0.0.1:<port>` (Windows).
  Generated sandbox client parses both.

- cron/scheduler.py: `argv = ["/bin/bash", str(path)]` hardcoded.  Use
  shutil.which("bash") so Windows (Git Bash via MinGit) works, with a
  readable error when bash is genuinely absent.

- 6 bare npm/npx spawn sites: tools_config.py x2, doctor.py, whatsapp.py
  (npm install + node version probe), browser_tool.py x2.  On Windows npm
  is npm.cmd / npx is npx.cmd (batch shims); subprocess.Popen(["npm", ...])
  fails with WinError 193.  shutil.which(...) returns the absolute .cmd
  path which CreateProcessW accepts because the extension routes through
  cmd.exe /c.  POSIX behaviour unchanged (shutil.which still returns the
  same path subprocess would resolve itself).

## HIGH fixes (silent misbehaviour on Windows)

- tools/environments/local.py get_temp_dir: hardcoded /tmp returned on
  Windows meant `_cwd_file = "/tmp/hermes-cwd-*.txt"`, which bash wrote
  via MSYS2's virtual /tmp but native Python couldn't open.  Result: cwd
  tracking silently broken — `cd` in terminal tool did nothing.  Windows
  branch now returns `%HERMES_HOME%/cache/terminal` with forward slashes
  (works in both bash and Python, guaranteed no spaces).

- tools/environments/local.py _make_run_env PATH injection: `/usr/bin not
  in split(":")` heuristic mangles Windows PATH (";" separator).  Gate
  the injection behind `not _IS_WINDOWS`.

- hermes_cli/gateway.py launch_detached_profile_gateway_restart: outer
  Popen + watcher-script Popen both used start_new_session=True, which
  Windows silently ignores.  Watcher stayed attached to CLI's console,
  died when user closed terminal after `hermes update`, left gateway
  stale.  Now branches through windows_detach_popen_kwargs() helper
  (CREATE_NEW_PROCESS_GROUP | DETACHED_PROCESS | CREATE_NO_WINDOW on
  Windows, start_new_session=True on POSIX — identical to main).

## MEDIUM fixes

- gateway/run.py /restart and /update handlers: hardcoded bash/setsid
  chain crashes on Windows when user triggers /update in-gateway.  Now
  has sys.platform=="win32" branch using sys.executable + a tiny
  Python watcher with proper detach flags.  POSIX path is unchanged.

- cli.py _git_repo_root: Git on Windows sometimes returns /c/Users/...
  style paths that break subprocess.Popen(cwd=...) and Path().resolve().
  Added _normalize_git_bash_path() helper that translates /c/Users,
  /cygdrive/c, /mnt/c variants to native C:\Users form.  POSIX no-op.
  _git_repo_root() now routes every result through it.

- cli.py worktree .worktreeinclude: os.symlink on directories failed
  hard on Windows (requires admin or Developer Mode).  Falls back to
  shutil.copytree with a warning log.

## Tests

- 29 new tests in tests/tools/test_windows_native_support.py covering:
  subprocess_compat helpers, TUI entry signal guards, kanban waitpid
  guard, code_execution TCP fallback source-level invariants, cron bash
  resolution, npm/npx bare-spawn lint per-file, local env Windows temp
  dir, PATH injection gating, git bash path normalization, symlink
  fallback, gateway detached watcher flags.

- One existing test assertion adjusted in test_browser_homebrew_paths:
  it compared captured Popen argv to the BARE `"npx"` literal; after the
  shutil.which() change argv[0] is the absolute path.  New assertion
  checks the shape (two items, second is `agent-browser`) rather than
  the exact first-item string.  Behaviour unchanged; test was too strict.

All 56 tests pass on Linux (30 from previous commits + 26 new).
267 tests from the affected files/dirs (browser, code_exec, local_env,
process_registry, kanban_db, windows_compat) all pass — zero regressions.
tests/hermes_cli/ (3909 pass) and tests/gateway/ (5021 pass) unchanged;
all pre-existing test failures confirmed unrelated via `git stash` re-run.

## What's still deferred (LOW priority)

- Visible cmd-window flashes on short-lived console apps (~14 sites) —
  cosmetic, needs a follow-up pass once we have user reports.
- agent/file_safety.py POSIX-only security deny patterns — separate
  hardening task.
- tools/process_registry.py returning "/tmp" as fallback — theoretical;
  reachable only when all env-var candidates fail.
This commit is contained in:
Teknium 2026-05-07 17:29:31 -07:00
parent b53bd12fe4
commit e93bfc6c93
15 changed files with 955 additions and 70 deletions

View file

@ -235,10 +235,27 @@ _call_lock = threading.Lock()
''' + _COMMON_HELPERS + '''\
def _connect():
"""Connect to the parent's RPC server via the transport it picked.
HERMES_RPC_SOCKET can be either:
- a filesystem path (POSIX Unix domain socket the default on
Linux and macOS)
- a string of the form ``tcp://127.0.0.1:<port>`` (Windows, where
AF_UNIX is unreliable the parent falls back to loopback TCP)
"""
global _sock
if _sock is None:
_sock = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM)
_sock.connect(os.environ["HERMES_RPC_SOCKET"])
endpoint = os.environ["HERMES_RPC_SOCKET"]
if endpoint.startswith("tcp://"):
# tcp://host:port (host is always 127.0.0.1 in practice — we
# only bind loopback server-side)
_host_port = endpoint[len("tcp://"):]
_host, _, _port = _host_port.rpartition(":")
_sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
_sock.connect((_host or "127.0.0.1", int(_port)))
else:
_sock = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM)
_sock.connect(endpoint)
_sock.settimeout(300)
return _sock
@ -988,8 +1005,22 @@ def execute_code(
# Use /tmp on macOS to avoid the long /var/folders/... path that pushes
# Unix domain socket paths past the 104-byte macOS AF_UNIX limit.
# On Linux, tempfile.gettempdir() already returns /tmp.
#
# Windows: Python 3.9+ added partial AF_UNIX support but the file-backed
# variant is flaky across Windows builds (requires Windows 10 1803+,
# still fails under some configurations, and the socket file can't live
# on the same temp drive as the script). Fall back to loopback TCP —
# same ephemeral port, same 1-connection listen queue, same serialized
# request/response framing. The generated client reads the transport
# selector from HERMES_RPC_SOCKET (path vs. ``tcp://host:port``).
_sock_tmpdir = "/tmp" if sys.platform == "darwin" else tempfile.gettempdir()
sock_path = os.path.join(_sock_tmpdir, f"hermes_rpc_{uuid.uuid4().hex}.sock")
_use_tcp_rpc = _IS_WINDOWS
if _use_tcp_rpc:
sock_path = None # not used on Windows; TCP endpoint stored below
rpc_endpoint = None # set after bind()
else:
sock_path = os.path.join(_sock_tmpdir, f"hermes_rpc_{uuid.uuid4().hex}.sock")
rpc_endpoint = sock_path
tool_call_log: list = []
tool_call_counter = [0] # mutable so the RPC thread can increment
@ -1008,10 +1039,24 @@ def execute_code(
with open(os.path.join(tmpdir, "script.py"), "w") as f:
f.write(code)
# --- Start UDS server ---
server_sock = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM)
server_sock.bind(sock_path)
os.chmod(sock_path, 0o600)
# --- Start RPC server ---
# Two transports:
# POSIX: AF_UNIX stream socket on sock_path, chmod 0600 for
# owner-only access. Filesystem permissions gate the socket.
# Windows: AF_INET stream socket on 127.0.0.1 with an ephemeral
# port. No filesystem permission story, but loopback-only bind
# means only the current user's processes (not remote) can
# connect. HERMES_RPC_SOCKET is set to ``tcp://127.0.0.1:<port>``
# which the generated client parses to pick AF_INET.
if _use_tcp_rpc:
server_sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
server_sock.bind(("127.0.0.1", 0)) # ephemeral port
_host, _port = server_sock.getsockname()[:2]
rpc_endpoint = f"tcp://{_host}:{_port}"
else:
server_sock = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM)
server_sock.bind(sock_path)
os.chmod(sock_path, 0o600)
server_sock.listen(1)
rpc_thread = threading.Thread(
@ -1053,7 +1098,7 @@ def execute_code(
# Allow vars with known safe prefixes.
if any(k.startswith(p) for p in _SAFE_ENV_PREFIXES):
child_env[k] = v
child_env["HERMES_RPC_SOCKET"] = sock_path
child_env["HERMES_RPC_SOCKET"] = rpc_endpoint
child_env["PYTHONDONTWRITEBYTECODE"] = "1"
# Ensure the hermes-agent root is importable in the sandbox so
# repo-root modules are available to child scripts. We also prepend
@ -1302,7 +1347,10 @@ def execute_code(
import shutil
shutil.rmtree(tmpdir, ignore_errors=True)
try:
os.unlink(sock_path)
# Only UDS has a filesystem socket to unlink; TCP sockets are
# freed by server_sock.close() above.
if sock_path:
os.unlink(sock_path)
except OSError:
pass # already cleaned up or never created