fix(scripts): fail loud on missing node/WS config; harden ota_convert_stock

ms_ws.py: read MS_WS/NODE_ID from the env or ota-release.env and raise a clear error if
neither has them, instead of silently defaulting to ws://localhost:5580 / node 9 (both
stale -- matter-server runs on the Pi, node ids are per-device). Rename drive9.py ->
drive_ac.py (the name encoded the stale node).

ota_convert_stock.sh: set -euo pipefail like its sibling scripts so a mid-sequence failure
(e.g. a failed chip-tool call during a live conversion) halts instead of silently
continuing; the informational greps + interactive teardown are || true-guarded.

Assisted-by: AI

Author: AndrewDemsDS

firmware/scripts/drive9.py firmware/scripts/drive_ac.pyfirmware/scripts/drive9.py renamed to firmware/scripts/drive_ac.py

@@ -3,7 +3,7 @@
 
 
 async def main():
-    nid = ms_ws.node_id(9)
+    nid = ms_ws.node_id()
     async with ms_ws.connect(heartbeat=30, hello_timeout=20) as (ws, _):
         seq = [
             ("SystemMode=Off", "1/513/28", 0),

firmware/scripts/ms_ws.py

@@ -2,31 +2,65 @@
 
 One async call() (send a command, await the reply whose message_id matches) and
 one async connect() context manager (open the WS, consume the server_info hello
-frame). Both read the matter-server WS endpoint from $MS_WS (default
-ws://localhost:5580/ws); node_id() reads the target node from $NODE_ID.
+frame). ws_url() and node_id() read $MS_WS / $NODE_ID, falling back to the same
+ota-release.env that ota-release.sh uses (so a bare `python3 <script>.py` targets
+the right Pi + node). There is NO silent default: matter-server runs on the Pi and
+the node id is per-device, so a wrong endpoint/node quietly drives nothing -- both
+raise a clear error instead (the old ws://localhost:5580 + node 9 defaults were
+stale and silently mistargeted).
 
-Factored out of mcli/mnodes/ota_go/drive9, which each previously carried their
+Factored out of mcli/mnodes/ota_go/drive_ac, which each previously carried their
 own copy of call() plus a hardcoded ws://localhost:5580/ws + node 9.
 """
 
 import os
+import re
 import json
 from contextlib import asynccontextmanager
 
 import aiohttp
 
-DEFAULT_WS = "ws://localhost:5580/ws"
+# ota-release.env sits next to this module; it's the source of truth for MS_WS + NODE_ID.
+_ENV_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "ota-release.env")
+
+
+def _env(key):
+    """$key from the environment, else the same key parsed from ota-release.env, else None.
+    The env always wins; the file is only a fallback for bare invocation."""
+    v = os.environ.get(key)
+    if v:
+        return v
+    try:
+        with open(_ENV_FILE) as f:
+            for line in f:
+                m = re.match(rf'\s*{key}\s*=\s*"?([^"#\n]+?)"?\s*(?:#.*)?$', line)
+                if m:
+                    return m.group(1)
+    except OSError:
+        pass
+    return None
 
 
 def ws_url():
-    """Matter-server WS endpoint ($MS_WS, default ws://localhost:5580/ws)."""
-    return os.environ.get("MS_WS", DEFAULT_WS)
+    """Matter-server WS endpoint from $MS_WS or ota-release.env (required, no default)."""
+    v = _env("MS_WS")
+    if not v:
+        raise SystemExit(
+            "MS_WS not set -- export it or set it in firmware/scripts/ota-release.env "
+            "(e.g. ws://<pi-host>:5580/ws). matter-server runs on the Pi, not localhost."
+        )
+    return v
 
 
-def node_id(default=9):
-    """Target node id ($NODE_ID if set, else the caller's default)."""
-    v = os.environ.get("NODE_ID")
-    return int(v) if v else default
+def node_id():
+    """Target node id from $NODE_ID or ota-release.env (required, no default)."""
+    v = _env("NODE_ID")
+    if not v:
+        raise SystemExit(
+            "NODE_ID not set -- export it or set it in firmware/scripts/ota-release.env "
+            "(per-device; e.g. kitchen AmebaZ2=14, ESP32=28)."
+        )
+    return int(v)
 
 
 async def call(ws, cmd, args, mid, timeout=180):

firmware/scripts/ota_convert_stock.sh

@@ -15,7 +15,11 @@
 #   IOT_SSID    your-iot-ssid          IOT_PW_FILE  ~/.iot_wifi_pw
 #   NODE        100 (target on chip-tool fabric)   PROV_NODE 1
 #   NEWVER      23   CURVER_MAX 22     CODE 34970112332
-set -o pipefail
+# Hardened like its sibling scripts: -e halts a mid-sequence failure (e.g. a failed chip-tool
+# write) instead of silently continuing; -u catches unset vars; pipefail propagates pipe failures.
+# Informational `... | grep | tail` filters (which legitimately match nothing) and the interactive
+# teardown are explicitly `|| true`-guarded so they don't trip -e.
+set -euo pipefail
 CHIP="${CHIP:-$HOME/ameba-dev/connectedhomeip}"
 CT="$CHIP/examples/chip-tool/out/host/chip-tool"
 OP="$CHIP/examples/ota-provider-app/linux/out/host/chip-ota-provider-app"
@@ -37,10 +41,10 @@ case "${1:-}" in
     echo ">>> Commissioning (bypass). Retry this subcommand if you see IM 0x0501 (transient)."
     "$CT" pairing code-wifi "$NODE" "$IOT_SSID" "$(cat "$IOT_PW_FILE")" "$CODE" \
       --bypass-attestation-verifier 1 --storage-directory "$STORE" 2>&1 | \
-      grep -iE "Commissioning complete|CHIP Error|IM Error|Timeout|success" | tail -6
+      grep -iE "Commissioning complete|CHIP Error|IM Error|Timeout|success" | tail -6 || true
     echo ">>> Identity + OTA Requestor check:"
     for a in vendor-id product-id software-version; do
-      "$CT" basicinformation read "$a" "$NODE" 0 --storage-directory "$STORE" 2>&1 | grep -iE "VendorID|ProductID|SoftwareVersion" | tail -1
+      "$CT" basicinformation read "$a" "$NODE" 0 --storage-directory "$STORE" 2>&1 | grep -iE "VendorID|ProductID|SoftwareVersion" | tail -1 || true
     done
     "$CT" descriptor read server-list "$NODE" 0 --storage-directory "$STORE" 2>&1 | grep -iE "42 \(Ota" && \
       echo ">>> OTA Requestor present. Run: $0 ota <VID> <PID>" || echo ">>> NO OTA Requestor (42) — use CH341A flash instead."
@@ -52,7 +56,7 @@ case "${1:-}" in
     python3 "$TOOL" extract "$OTA_SRC" "$WORK/payload.bin" >/dev/null 2>&1 || die "extract failed"
     python3 "$TOOL" create -v "$VID" -p "$PID" -vn "$NEWVER" -vs "${NEWVER}.0" -mi 1 -ma "$CURVER_MAX" \
       -da sha256 "$WORK/payload.bin" "$TGT" >/dev/null 2>&1 || die "create failed"
-    python3 "$TOOL" show "$TGT" | grep -iE "Vendor Id|Product Id|Version:"
+    python3 "$TOOL" show "$TGT" | grep -iE "Vendor Id|Product Id|Version:" || true
     if printf '%s\n' "$@" | grep -qx -- --dry-run; then
       echo ">>> --dry-run: .ota repackaged + header verified (host-only); skipping provider/commission/OTA."
       exit 0
@@ -61,13 +65,13 @@ case "${1:-}" in
     rm -f /tmp/ota_prov.kvs
     "$OP" --discriminator 22 --secured-device-port 5560 --KVS /tmp/ota_prov.kvs --filepath "$TGT" > "$WORK/provider.log" 2>&1 &
     PP=$!; sleep 4
-    "$CT" pairing onnetwork "$PROV_NODE" 20202021 --storage-directory "$STORE" 2>&1 | grep -iE "complete|Error" | tail -2
+    "$CT" pairing onnetwork "$PROV_NODE" 20202021 --storage-directory "$STORE" 2>&1 | grep -iE "complete|Error" | tail -2 || true
     "$CT" accesscontrol write acl "[{\"fabricIndex\":1,\"privilege\":5,\"authMode\":2,\"subjects\":[112233],\"targets\":null},{\"fabricIndex\":1,\"privilege\":3,\"authMode\":2,\"subjects\":null,\"targets\":null}]" "$PROV_NODE" 0 --storage-directory "$STORE" >/dev/null 2>&1
     "$CT" otasoftwareupdaterequestor write default-otaproviders "[{\"fabricIndex\":1,\"providerNodeID\":$PROV_NODE,\"endpoint\":0}]" "$NODE" 0 --storage-directory "$STORE" >/dev/null 2>&1
-    "$CT" otasoftwareupdaterequestor announce-otaprovider "$PROV_NODE" 0 0 0 "$NODE" 0 --storage-directory "$STORE" 2>&1 | grep -iE "Status=0x0|Error" | tail -1
+    "$CT" otasoftwareupdaterequestor announce-otaprovider "$PROV_NODE" 0 0 0 "$NODE" 0 --storage-directory "$STORE" 2>&1 | grep -iE "Status=0x0|Error" | tail -1 || true
     echo ">>> Announced. Watching transfer (Ctrl-C when you see ApplyUpdateRequest)..."
-    timeout 300 tail -f "$WORK/provider.log" | grep --line-buffered -iE "QueryImage|UpdateAvailable|BlockAckEOF|ApplyUpdateRequest"
-    kill "$PP" 2>/dev/null
+    timeout 300 tail -f "$WORK/provider.log" | grep --line-buffered -iE "QueryImage|UpdateAvailable|BlockAckEOF|ApplyUpdateRequest" || true
+    kill "$PP" 2>/dev/null || true
     echo ">>> On ApplyUpdateRequest the unit reboots into our firmware. Then press 77 and add to HA with code $CODE."
     ;;
   *) echo "usage: $0 {commission | ota <VID> <PID> [--dry-run]}  (see firmware/docs/12-ota-convert-stock-unit.md)";;

firmware/scripts/ota_go.py

@@ -3,7 +3,7 @@
 
 
 async def main():
-    nid = ms_ws.node_id(9)
+    nid = ms_ws.node_id()
     # wait for node available
     for attempt in range(20):
         try:
@@ -26,7 +26,7 @@ async def main():
                         )
                         print("update_node result:", json.dumps(r)[:400])
                     return
-        except Exception as e:
+        except Exception:
             pass
         await asyncio.sleep(10)
     print(f"node{nid} never became available")