From f720a16f8e4ac538e6b880960fd411357114656d Mon Sep 17 00:00:00 2001 From: user Date: Fri, 31 Jul 2026 09:14:41 +0200 Subject: rpc: stop lying to the driver about renames, modals and teardown MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Five defects found while an agent drove a long RE session over the socket. Each one was reproduced on a live spawned pane first (an in-process pilot would not have shown any of them), then fixed: pane stop truncated the save. `stop` asked the app to quit, slept 400 ms, then unconditionally killed the pane. Quitting runs App.on_unmount, which writes every dirty database; a 90 MB .i64 takes tens of seconds, so the kill landed mid-write and a whole session's annotations went to /dev/null with a cheerful {"stopped": [...]} on stdout. Now it waits for the pane to actually exit (--timeout, default 600 s) and only force-kills on timeout, saying so. The quit verb bypassed the dirty check. It called app.exit() directly rather than the path a human gets, so the "unsaved changes" logic never ran. It now routes through _on_quit_choice and reports {saving, dirty}. Naming a function start from the listing never reached the function index. `goto ` puts the cursor on the address token, so `n` takes the name-an-address path, which called bump_items() but left FunctionIndex holding the old name. Result: the rename response snapshot showed the section label, and functions()/names()/the palette all reported the rename had not happened — so a driver that trusts its readbacks redoes work it already did. Twice, in the session that prompted this. _do_name_addr now updates the index, the nav stack and the table cell when the address is a function start. A stripped binary with no entry function started up *inside a modal*. _auto_land pushed the symbol palette when main() was missing, while ping still answered ready:true. Every keystroke an RPC driver injected went into the palette's search box and was silently swallowed. It now lands on the first function instead and hints at Ctrl+N. Verbs that inject keystrokes now refuse when a modal is on top, naming it, instead of failing with "'goto' prompt did not open (word under cursor?)" — a message that blamed the cursor for what was always a focus problem. Also: `drive raw` passes k=v values through as strings, so `view lines=8` died with "'<' not supported between instances of 'int' and 'str'". Numeric params are now coerced centrally rather than at each call site. tests/test_scenarios.py: 212 passed, 0 failed. --- idatui/pane.py | 22 +++++++++++++++++++++- 1 file changed, 21 insertions(+), 1 deletion(-) (limited to 'idatui/pane.py') diff --git a/idatui/pane.py b/idatui/pane.py index 1a46a4f..37a5f0c 100644 --- a/idatui/pane.py +++ b/idatui/pane.py @@ -248,18 +248,29 @@ def stop(args) -> int: if not rows: print("error: no matching pane (need --sock or --pane)", file=sys.stderr) return 2 + killed: list[str] = [] for r in rows: sock, pane = r.get("sock"), r.get("pane") + quit_ok = False if sock and os.path.exists(sock): try: # ask it to quit gracefully first with RpcClient(sock) as c: c.call("quit") - time.sleep(0.4) + quit_ok = True except (OSError, RpcError, ConnectionError): pass + # Wait for the pane to actually go away. Quitting runs App.on_unmount, + # which writes every dirty database; a 90 MB .i64 takes tens of seconds. + # Killing the pane on a fixed short sleep truncated that save and + # silently destroyed the session's work, so block on the real signal. + if pane and quit_ok: + deadline = time.monotonic() + float(args.timeout) + while time.monotonic() < deadline and _pane_alive(pane): + time.sleep(0.25) if pane and _pane_alive(pane): subprocess.run(["tmux", "kill-pane", "-t", pane], capture_output=True) + killed.append(pane) if sock: try: os.unlink(sock) @@ -271,6 +282,12 @@ def stop(args) -> int: out = {"stopped": [r.get("sock") or r.get("pane") for r in rows]} if reaped: out["reaped_workers"] = reaped + if killed: + # Only ever reached on timeout: say so, because it means a save may have + # been cut short rather than "clean teardown". + out["force_killed"] = killed + out["warning"] = (f"pane(s) did not exit within {args.timeout}s and were " + "killed; unsaved database changes may be lost") print(json.dumps(out)) return 0 @@ -335,6 +352,9 @@ def main(argv: list[str]) -> int: st = sub.add_parser("stop", help="graceful quit + kill the pane") st.add_argument("--sock") st.add_argument("--pane") + st.add_argument("--timeout", type=float, default=600.0, + help="seconds to wait for the pane to exit (it saves dirty " + "databases on the way out) before force-killing it") st.set_defaults(fn=stop) ls = sub.add_parser("list", help="list tracked panes") -- cgit v1.3.1-sl0p