dohertj2/mxaccess
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
92 Commits 1 Branch 0 Tags
b1a5f5ff1e019f55bf9ffe3adcb90bbafd4d46f0
Commit Graph

17 Commits

Author SHA1 Message Date
Joseph Doherty 1de049e114 [F2] mxaccess-rpc: NTLM verify_signature (server-to-client) with constant-time MAC compare
rust / build / test / clippy / fmt (push) Has been cancelled
Details 
Closes F2. Structural port from [MS-NLMP] §3.4.4 — same shape as
the existing sign path but uses the server-to-client sub-keys
(`SealKey_S→C` / `SignKey_S→C`) derived alongside the client-to-
server pair at the end of create_type3.

NtlmClientContext gained four new fields populated during
create_type3:
  - server_signing_key
  - server_sealing_key
  - server_sealing_state (independent RC4 stream)
  - server_sequence (independent counter)

The S→C key derivation already existed in auth.rs (the seal_key /
sign_key helpers take a client_mode flag); F2 plumbs them into a
new verify_signature(message, signature) method.

The verify path:
  1. Validates signature.len() == 16 + leading version word 0x01.
  2. Reads trailing seq num, compares against self.server_sequence
     (mismatch ⇒ InvalidSignature, no state change).
  3. Computes expected_mac = HMAC_MD5(server_signing_key,
     seq || message)[0..8] then RC4 transform.
  4. Constant-time compares expected_mac against wire bytes 4..12
     via subtle::ConstantTimeEq.
  5. On success: commits cipher-state advance + ++server_sequence.
     On failure: re-derives RC4 from server_sealing_key and skips
     past server_sequence × 8 keystream bytes to restore the
     pre-verify position — caller can retry.

New dep `subtle = "2"` (workspace-internal to mxaccess-rpc) for
the timing-oracle-safe MAC compare.

6 new tests:
  - verify_signature_round_trip_against_sign (3-message sequence
    via paired_authed_context helper that aliases server-side keys
    onto client-side for self-validating round-trip)
  - verify_signature_rejects_corrupted_mac (with
    server_sequence-non-advance assertion)
  - verify_signature_rejects_wrong_sequence_number
  - verify_signature_rejects_wrong_version_field
  - verify_signature_rejects_wrong_length
  - verify_signature_before_authenticate_errors

mxaccess-rpc 188 → 194 tests; default-feature clippy clean.

The "awaiting wire-fixture capture" step listed in F2's prior
status note is no longer a hard prerequisite — [MS-NLMP] §3.4.4
fully defines the algorithm and the round-trip tests prove the
encoder/decoder pair is internally consistent. A captured
StatusReceived frame would still validate byte-parity vs a real
NmxSvc.exe signer, but that's future verification work; the
structural port ships unblocked.

design/followups.md F2 moved to Resolved.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-06 03:30:48 -04:00
Joseph Doherty 9496322712 [F27] mxaccess-asb-nettcp: constant-time DH mod_exp via crypto-bigint::DynResidue
rust / build / test / clippy / fmt (push) Has been cancelled
Details 
Closes F27 per option (b) of its resolve criterion: fixed-width
U2048 DH backend using crypto-bigint's Montgomery-form residue
arithmetic.

New auth.rs::constant_time_mod_exp(base, exp, modulus) wrapper
preserves the BigUint-in-BigUint-out API of the existing byte
helpers; the actual square-and-multiply chain runs in Montgomery
form. Both DH call sites swap to the wrapper:
  - AsbAuthenticator::new line 179 (public-key generation)
  - crypto_key line 354 (shared-secret derivation)

DH private exponent timing-leak resistance is the goal: the .NET
reference's BigInteger.ModPow is also non-constant-time, so we
were at parity but not at the long-term Rust target. With this
fix the production path no longer leaks the bit-pattern of the
long-lived DH private key through power/timing side channels.

DynResidueParams::new requires an odd modulus (Montgomery form's
only restriction). Production DH primes are always odd
(`MX_ASB_DH_PRIME = 1552...7919` on this host's registry).
CryptoParameters::DEFAULT_PRIME_TEXT — the test-fixture default
inherited from AsbRegistry.cs:66 — ends in 4 (even), which is
mathematically unsound for DH but kept for parity with the .NET
default. For that case the wrapper falls back to BigUint::modpow,
preserving the wire bytes (modular exp is a pure function of
inputs).

Wire-byte parity verified two ways:
1. Unit fixture test
   `auth.rs::deterministic_hmac_matches_dotnet_fixture` — byte-equal
   to captured .NET output for the full DH → PBKDF2 → AES-CBC chain.
   Continues to pass.
2. Live: Connect handshake against the local AVEVA install
   completes with apollo:V2 lifetime, proving MxDataProvider
   accepts the constant-time-derived public key and the
   shared-secret-based AuthenticateMe.

Workspace deps:
  - crypto-bigint = "0.5" added to [workspace.dependencies] and
    mxaccess-asb-nettcp/Cargo.toml.
  - num-bigint retained for decimal-string parsing + .NET-LE byte
    conversion (crypto-bigint has neither).

Closes the "review.md MAJOR finding" originally flagged at
design/30-crate-topology.md:269-274.

design/followups.md: F27 moved to Resolved.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-06 03:16:33 -04:00
Joseph Doherty cf9dbaf568 [F6] mxaccess-rpc: ComObjRefProvider port via windows-rs (CoMarshalInterface)
rust / build / test / clippy / fmt (push) Has been cancelled
Details 
New module crates/mxaccess-rpc/src/com_objref_provider.rs gated on
cfg(all(windows, feature = "windows-com")). Pulls windows = "0.59"
(features Win32_Foundation + Win32_System_Com +
Win32_System_Com_Marshal + Win32_System_Com_StructuredStorage +
Win32_System_Memory) as an optional dep behind the existing
windows-com feature; default footprint stays slim.

Public API mirrors ComObjRefProvider.cs 1:1: MarshalContext enum
(InProcess / Local / DifferentMachine wrapping the MSHCTX_* newtype
constants), clsid_from_prog_id, marshal_activated_iunknown_objref
(activates via CoCreateInstance with INPROC | LOCAL | REMOTE then
marshals), marshal_iunknown_objref (uses IUnknown::IID),
marshal_interface_objref (CoMarshalInterface over an HGlobal-backed
IStream).

All `unsafe` is internal to the module — public API exposes only
typed Rust values (Vec<u8>, GUID, ProviderError), no raw pointers /
HRESULTs / lifetime-bound interface pointers leak. Each unsafe block
carries an inline SAFETY comment naming the invariants being upheld.

Per-thread COM init via thread-local OnceLock<()>: lazy
CoInitializeEx(MULTITHREADED) on first call; S_FALSE (already
initialised) and RPC_E_CHANGED_MODE (thread is STA) treated as
success — matches the .NET runtime's tolerant apartment behaviour.

ProviderError enumerates the four documented failure modes plus the
apartment-init pre-check: UnknownProgId / ActivationFailed /
MarshalFailed / GlobalLockFailed / ApartmentInitFailed.

4 offline tests: MarshalContext → MSHCTX_* mapping, ensure_apartment
idempotence, clsid_from_prog_id returns UnknownProgId for fake
ProgIDs, marshal_activated short-circuits at the resolution stage.

1 live test (#[ignore], gated on MX_LIVE): activates the real
NmxSvc.NmxService, marshals the proxy's IUnknown via
CoMarshalInterface, then parses the resulting blob via
ComObjRef::parse and asserts non-zero OXID + IPID. Passes against
the AVEVA install on this host.

Workspace tests: mxaccess-rpc went 179 → 183 (+4). All other crates
unchanged.

Unblocks F12 (NmxClient::create — the auto-resolving
COM-activation factory): the underlying primitive
(marshal_activated_iunknown_objref) now exists; remaining work is
threading the windows-com feature through mxaccess-nmx and chaining
ComObjRef::parse → resolve_oxid_with_managed_ntlm_packet_integrity →
RemQueryInterface. design/followups.md F12 updated with a revised
"Resolves when" reflecting that F6's blocker is gone.

Closes F6 in design/followups.md.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 22:11:33 -04:00
Joseph Doherty 41f2d4c0f2 [F14] mxaccess-galaxy: tiberius-backed SQL Resolver + UserResolver
rust / build / test / clippy / fmt (push) Has been cancelled
Details 
New module crates/mxaccess-galaxy/src/sql_resolver.rs (~480 LoC) gated
behind the existing galaxy-resolver Cargo feature. Adds SqlTagResolver
+ SqlUserResolver, both constructed via from_ado_string(&str)
accepting the same connection-string shape the .NET reference uses by
default (Server=localhost;Database=ZB;Integrated Security=True;
Encrypt=False;TrustServerCertificate=True). Integrated Security=True
resolves to Windows auth via tiberius's winauth feature.

Each top-level call (resolve / browse / resolve_by_guid /
resolve_by_name) opens a fresh Client<Compat<TcpStream>> and drops it
on return — matches the .NET `await using` lifecycle at
GalaxyRepositoryTagResolver.cs:93-95. tiberius's Client::query only
accepts positional @P1..@PN placeholders (delegates to sp_executesql);
the canonical RESOLVE_SQL / BROWSE_SQL / USER_BY_GUID_SQL /
USER_BY_NAME_SQL constants are rewritten once-per-process via
OnceLock<String> (@objectTagName → @P1, etc.). The unrewritten
constants stay byte-identical with the .NET reference for ad-hoc
diagnostic copy/paste.

read_metadata mirrors ReadMetadata (cs:149-165) byte-by-byte: signed
smallint → i16 widened to u16 for platform/engine/object IDs (matches
the .NET checked((ushort)reader.GetInt16(N)) shape), int → i32
checked-cast to i16 for property_id, nullable nvarchar for
primitive_name. read_user_profile mirrors ReadProfile (cs:76-85)
including the roles_text blob → parse_role_blob round-trip.

Deps added (gated): tiberius 0.12 (default-features = false; tds73 +
rustls + winauth — no chrono / rust_decimal pulled), tokio-util's
compat feature for the futures-rs ↔ tokio AsyncRead bridge,
futures-util for TryStreamExt::try_next. Default-feature build still
pulls only mxaccess-codec + async-trait + thiserror + uuid (slim
foot-print preserved per the design doc's intent).

New `live` feature on this crate (`live = ["galaxy-resolver"]`) for
parity with the workspace pattern.

11 offline unit tests pin: SQL named→positional rewriting (no @named
left, @P1/@P2/@P3 present), line-count preserved, ado-string
acceptance (default Galaxy shape parses, garbage rejected), input
validation (max_rows=0 rejected, empty LIKE rejected, empty user_name
rejected, all checked before connect attempt).

Two #[cfg(feature = "live")] #[ignore]'d tests round-trip against a
real Galaxy DB (gated on MX_LIVE + MX_GALAXY_DB env vars per
tools/Setup-LiveProbeEnv.ps1). Live verification on this host:
live_resolve_test_child_object_test_int and
live_browse_test_child_object both pass against the local AVEVA
install — TestChildObject.TestInt resolves with mx_data_type=2
(Int32), is_array=false.

Closes F14 in design/followups.md.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 21:54:43 -04:00
Joseph Doherty 3b09297b27 [M5] live-probe iteration 1 — major wire-byte reconciliation fixes 
First live-test cycle against AVEVA on this box. Comparing the .NET
probe's `--dump-messages` XML output against our NBFX-encoded
envelope surfaced six structural bugs in the F25 envelope/operations
layer. All fixed; tests passing (702 workspace).

Fixes (all backed by the .NET dump as ground truth):

1. **`mustUnderstand` attribute name** — NBFS dict id was 116
   (`MustUnderstand`, capital-M, a different SOAP token); SOAP 1.2
   spec uses lowercase `mustUnderstand` at id 0. Sending the wrong
   one triggered a WCF parse fault that surfaced as TCP RST.

2. **Missing `<a:MessageID>` header** — WCF's default binding
   requires MessageID for two-way operations. We now auto-generate
   `urn:uuid:<v4>` per envelope via a small inline `make_random_uuid_v4`
   helper (no `uuid` crate dep).

3. **Missing `<a:ReplyTo>` anonymous header** — WCF's
   BinaryMessageEncoder always emits `<a:ReplyTo><a:Address>...
   addressing/anonymous</a:Address></a:ReplyTo>` for two-way ops.

4. **ConnectionValidator field names + namespace** — we were
   emitting PascalCase `<ConnectionId>` etc. .NET's WCF
   DataContractSerializer uses the private backing-field names
   (`<connectionIdField xmlns="...ASBContract">guid</connectionIdField>`)
   per `[DataMember(Name = "fooField")]`. Added the
   `xmlns:i="...XMLSchema-instance"` declaration WCF emits
   alongside (even when no `i:nil` is used). Decoder now accepts
   both PascalCase (legacy tests) and DataContract field names.

5. **`<ASBIData>` over-wrapping** — we were emitting
   `<Items><ASBIData>{bytes}</ASBIData></Items>`. .NET's
   `AsbDataCustomSerializer.WriteStartObject` (`AsbContracts.cs:
   1561-1572`) REPLACES the field's outer element with `<ASBIData>`
   directly — there's no `<Items>` wrapper on the wire. Fixed by
   collapsing `BodyField::AsbiDataElement` to emit just `<ASBIData>`
   without the named outer element. The `name` field is retained
   for self-documentation.

6. **`collect_asbidata_payloads` API** — was keyed by field name
   (`Status` / `Values`); now positional (`payloads[0]`,
   `payloads.get(1)`) since the wrapper element is gone. All seven
   response decoders updated.

Plus tooling for the live-probe loop:
* `tools/Get-AsbPassphrase.ps1` — DPAPI loader that auto-discovers
  the solution name + reads the sharedsecret + decrypts it. Sets
  $env:MX_ASB_PASSPHRASE / MX_ASB_HOST / MX_ASB_VIA / MX_LIVE.
  Lowercase via-host (WCF SMSvcHost is case-sensitive on the URL
  host segment).
* `examples/asb-preamble-probe.rs` — diagnostic that connects,
  runs the preamble, captures the PreambleAck, then sends a
  synthetic ConnectRequest and dumps both directions as hex. Used
  to bisect the wire-byte deltas above.
* `examples/asb-subscribe.rs` port default fixed (5074 → 808 —
  WCF's NetTcpPortSharing/SMSvcHost listener confirmed via
  Get-NetTCPConnection).

**Status**: preamble + PreambleAck round-trip works end-to-end
against the live AVEVA install (verified via probe). The
post-preamble Connect SOAP envelope still gets TCP RST'd — the six
structural fixes above are necessary but not yet sufficient. Next
iteration needs binary wire capture (Wireshark + Npcap loopback,
or a TCP-relay middleman) to compare the .NET probe's BinaryMessageEncoder
output byte-for-byte with ours and find the remaining delta(s).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 15:06:48 -04:00
Joseph Doherty 8a0f92b6bc [M5] mxaccess: F26 step 1 — AsbTransport bridges AsbClient into Transport trait 
First slice of F26. Bridges F25's working AsbClient into the M0
`mxaccess::Transport` trait that Session uses to discriminate
operations across NMX and ASB transports.

API additions:
* `mxaccess::AsbTransport<T>` — generic over the same
  AsyncRead+AsyncWrite+Unpin+Send+Sync+'static bound that AsbClient
  takes. Owns an AsbClient and exposes it via `client_mut()` /
  `into_client()`.
* `impl Transport for AsbTransport<T>`:
  - `capabilities()` — `buffered_subscribe = false`,
    `activate_suspend = false`, `operation_complete_frame = false`
    per `design/60-roadmap.md` M5 (no NMX-specific extensions on
    ASB).
  - `kind()` — `TransportKind::Asb`.

Path-dep wiring: `mxaccess` now imports `mxaccess-asb` +
`mxaccess-asb-nettcp` directly.

Compile-time `Send + Sync + 'static` assertion guards the
trait-bound contract.

2 new tests:
* `asb_transport_kind_is_asb`.
* `asb_transport_capabilities_disable_buffered_and_activate_suspend`.

Stubbed for F26 step 2:
* `Session::connect_asb` constructor that owns TCP open +
  preamble + DH handshake orchestration.
* Operation routing that maps ASB types (ItemStatus, RuntimeValue)
  back to mxaccess types (MxStatus, DataChange, MxValue).

Stubbed for F26 step 3:
* Subscription routing — Session::subscribe on ASB needs F25
  subscription operations (CreateSubscription / AddMonitoredItems
  / Publish), which are not yet implemented.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 11:57:20 -04:00
Joseph Doherty 25dbd8d3bd [M5] mxaccess-asb: F25 step 1 — SOAP envelope codec 
First slice of F25. Provides the building blocks the per-operation
request/response codecs and the network loop will compose:

* `actions` module — IASBIDataV2 action strings (all 14 operations,
  verbatim from `AsbContracts.cs:14-58`).
* `ConnectionValidator` — SOAP header struct mirroring
  `AsbContracts.cs:65-117`. `from_signed(&SignedValidator)` converts
  F23's MAC + IV to base64 for the wire, matching .NET's
  `BinaryWriter`-via-`XmlSerializer` shape.
* `SoapEnvelope` + `encode_envelope` — assembles the NBFX token
  stream: `s:Envelope` → `s:Header` → `a:Action s:mustUnderstand="1"`
  → optional `h:ConnectionValidator` → `s:Body` → caller-supplied
  body tokens. Uses static-dictionary IDs for the SOAP/WS-Addressing
  tokens via F22's `lookup_static`.
* `decode_envelope` — pulls action + validator + body tokens back
  out of received bytes. Tolerant of header ordering.
* Mixed-endian GUID format/parse (`format_uuid` / `parse_uuid`) that
  mirrors .NET's `Guid.ToString("D")` byte order so connection-id
  round-trip matches the wire exactly.

9 new unit tests cover:
* Round-trip with and without validator.
* `from_signed` base64 encoding of MAC + IV.
* `format_uuid` produces the correct .NET-mixed-endian hex string.
* GUID round-trip through string formatter.
* Action string presence in the encoded byte stream.
* Decoder tolerance of envelopes without an Action header.
* Validator round-trip through full encode → decode.
* Lint-style guard that all 14 action constants are URIs ending `In`.

Stubbed for next F25 iteration: per-operation request/response
struct codecs (`ConnectRequest`, `RegisterItemsRequest`, etc.) +
`AsbClient` network loop.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 11:16:22 -04:00
Joseph Doherty ed17c07c10 [M5] mxaccess-asb-nettcp: M5 plan + F19 deps + F23 auth crypto port 
F18 plans M5 as 9 sub-followups (F18-F26 + F27 constant-time DH) per
design/dependencies.md:73-89. Wave-1 streams F20-F23+F24 are parallel-safe
after F19 (workspace deps). F25 (ASB client) is sequential after the
framing/encoder streams. F26 (Session over AsbTransport) is sequential
after F25.

F19 — workspace deps for the M5 crypto + framing surface: hmac, md-5,
sha1, sha2, aes, cbc, pbkdf2, flate2, rand, num-bigint, num-traits,
num-integer, quick-xml, tokio-util, zeroize. Pinned to the digest 0.10 /
cipher 0.4 generation matching mxaccess-rpc.

F23 — ports `AsbSystemAuthenticator.cs` (167 LoC) to
`mxaccess-asb-nettcp::auth`. Wire-byte parity points: .NET BigInteger
little-endian two's-complement byte order with optional 0x00 sign-byte
suffix; AES-128-CBC with PKCS7 padding; PBKDF2-SHA1 1000 iterations
over `Convert.ToBase64String(crypto_key)` with ASCII salt
"ArchestrAService"; deflate-then-AES (Baktun) vs raw-AES (Apollo)
selected by `:V2` lifetime suffix; HMAC-MD5/SHA1/SHA512 negotiated per
`AsbSolutionCryptoParameters.HashAlgorithm` (with `force_hmac=true`
fallback to HMAC-SHA1 for unrecognised algorithms).

13 unit tests cover the cryptographic primitives + DH peer agreement +
.NET byte-order round-trip + Apollo lifetime dispatch.

F27 — filed for the `num-bigint` → `crypto-bigint::BoxedUint` swap once
the latter exposes a stable heap-allocated `pow_mod`. Currently at
parity with the .NET reference (also not constant-time).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 10:36:15 -04:00
Joseph Doherty a31237d1d0 [M4] mxaccess: Subscription impls Stream<Item = DataChange> (resolves F15) 
F15 step 2/2 lands the per-Subscription routing on top of step 1's
broadcast layer. Subscription is now a working data-change stream.

Subscription type
- Now impls futures_util::Stream<Item = Result<DataChange, Error>>
  via tokio_stream::wrappers::BroadcastStream + a per-message filter.
- No longer Clone (broadcast::Receiver isn't Clone). Consumers that
  want fanout subscribe twice or share via Arc<Mutex<...>>.
- Holds the broadcast::Receiver subscribed BEFORE AdviseSupervisory
  fires — guarantees no updates between advise and stream-creation
  are dropped.
- pending VecDeque buffers records from the current message so each
  poll_next yields at most one DataChange (Stream contract).

Filter logic (records_to_data_changes, mirrors cs:333-343)
- 0x32 SubscriptionStatus: keep when
  msg.item_correlation_id == subscription.correlation_id; drop
  otherwise.
- 0x33 DataUpdate: keep ALL — codec exposes no per-record correlation
  field, and the .NET filter only checks item_correlation_id (which
  0x33 doesn't carry), so DataUpdates fan out to every active
  subscription. Matches .NET behavior verbatim.
- Records with value: None drop silently (mirrors evt.Record.Value
  is null filter at cs:337).
- BroadcastStream Lagged(n) maps to Error::Configuration with the
  lag count in the detail string.

Helpers
- filetime_to_system_time(i64) -> SystemTime: inverse of
  system_time_to_filetime; saturates at Unix epoch for FILETIMEs
  before 1970 since SystemTime can't portably represent pre-epoch.
- record_to_data_change(record, reference) -> Option<DataChange>:
  builds DataChange from one record, returns None for unparseable
  value (the codec couldn't decode the wire kind).
- Status currently hardcoded to MxStatus::DATA_CHANGE_OK (mirrors
  NmxSubscriptionRecord.ToDataChangeStatus at NmxSubscriptionMessage.cs:22-25
  which the .NET reference itself stubs to the OK constant).

Cargo.toml additions: futures-util (workspace) + tokio-stream (0.1
with sync feature for BroadcastStream).

Tests (5 new in mxaccess; total 40)
- subscription_stream_yields_data_change_for_matching_correlation:
  build a 0x32 SubscriptionStatus with one Int32 record and the
  subscription's correlation id, inject through test_inject_sender,
  observe the DataChange (reference, value, quality match) on the
  Stream.
- subscription_stream_filters_out_mismatched_correlation_for_status:
  inject 0x32 with wrong correlation id, assert the stream stays
  pending (timeout-as-success).
- subscription_stream_keeps_data_update_regardless_of_correlation:
  inject 0x33 DataUpdate with one Int32 record (no correlation
  field on the message); stream still yields the DataChange.
- filetime_to_system_time_round_trip: build a SystemTime with .005s
  precision, round-trip through both helpers, assert equality.
- filetime_to_system_time_pre_unix_epoch_saturates: FILETIME 0 (year
  1601) → SystemTime::UNIX_EPOCH (saturating clamp).

design/followups.md: F15 moved to Resolved with both step commits
referenced. Open list: 9 items (was 10).

Test count delta: 511 -> 516 (+5). All four DoD gates green.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 09:45:16 -04:00
Joseph Doherty 12cb10c3a1 [M4] mxaccess: Session::connect_nmx + write_value + shutdown (wave 1 main) 
First working M4 wave 1 slice. Adds session.rs with the connect /
write / shutdown path on top of NmxClient + Resolver, plus a tokio
test that exercises a full round-trip against a hand-rolled server.
Read, subscribe, recovery, and the long-lived connection task land
in wave 2.

Architecture
- Session holds Arc<SessionInner>; SessionInner wraps NmxClient
  behind a tokio::sync::Mutex. All RPC ops serialize on that mutex.
  Wave 2 will replace it with an mpsc::channel<Op> + dispatcher task
  per design/70-risks-and-open-questions.md R15 (drop-time async
  cleanup hazards).
- ensure_connected gate stops post-shutdown ops with
  Connection::EngineNotRegistered. Shutdown is idempotent via
  AtomicBool::swap.
- Manual Debug impl on SessionInner — neither dyn Resolver nor
  NmxClient impl Debug.

Public API
- Session::connect_nmx(addr, options, ntlm, service_ipid, resolver,
  recovery): validates the policy, opens NmxClient, runs
  RegisterEngine2 (no callback yet — wave 2), optionally configures
  heartbeat. Returns Error::Connection on non-zero HRESULT.
- Session::write_value(reference, value: WriteValue): resolves the
  tag through the configured Resolver, dispatches NmxClient::write.
- Session::resolve_write_kind / resolve_tag: convenience accessors.
- Session::shutdown_nmx: calls UnregisterEngine, idempotent.

Error mapping
- map_nmx / map_transport / map_resolver bridge the inner crate
  errors into the public Error enum. NonZeroHresult → InvalidArgument
  with the hex code; transport Fault → Status-shaped error;
  ResolverError::NotFound → Galaxy { reason: "tag not found: ..." }.
- All three matchers handle their #[non_exhaustive] sources with a
  generic catch-all so future variants don't silently break the map.

Tests (8 new in mxaccess; total mxaccess: 19)
- write_value round-trip via in-memory StaticResolver + hand-rolled
  unauthenticated DCE/RPC server.
- write_value propagates resolver not-found → Galaxy error.
- write_value propagates non-zero HRESULT → InvalidArgument.
- shutdown is idempotent (second call is a no-op).
- write after shutdown returns EngineNotRegistered.
- resolve_tag and resolve_write_kind work without RPC.
- envelope-kind constants used by Session match codec exports
  (sanity guard against codec rename).

mxaccess-nmx: WriteValue now re-exported at crate root.
mxaccess: deps gained mxaccess-nmx/galaxy/rpc + tokio + tracing,
plus async-trait as a dev-dep for the test resolver impl.

Test count delta: 479 -> 487 (+8). All four DoD gates green.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 09:01:44 -04:00
Joseph Doherty baea6eaa41 [M3] mxaccess-galaxy: GalaxyUserProfile + UserResolver trait + role-blob 
Lands the user-resolver half of M3 stream A. Pure-Rust foundation —
the tiberius-backed SQL impl is logged as F14 and stays gated behind
the existing galaxy-resolver Cargo feature.

New
- role_blob.rs (~270 LoC, 12 tests including a garbage-between-roles
  edge case) — port of ParseRoleBlob (cs:87-133). Sliding-window scan
  over hex-decoded UTF-16LE bytes; rejects non-printable code units;
  case-insensitive dedup. Pure function, no I/O.
- user.rs (~290 LoC, 8 tests including 4 tokio-driven InMemoryUserResolver
  cases) — GalaxyUserProfile (port of cs:5-11) + from_columns helper
  bridging into role_blob + UserResolver async trait + UserResolverError
  with NotFound / Backend variants.
- sql.rs additions: USER_SELECT_SQL + USER_BY_GUID_SQL + USER_BY_NAME_SQL
  constants (port of cs:135-148). Inline concatcp! macro composes the
  base SELECT with each WHERE clause at compile time without pulling
  const_format.

Cargo.toml: added uuid (Galaxy user_guid is a uniqueidentifier).

design/followups.md: added F14 (P2) for the tiberius-backed SQL impl
behind the galaxy-resolver feature.

Test count delta: 427 -> 446 (+19; mxaccess-galaxy 30 -> 49). All four
DoD gates green.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 08:26:24 -04:00
Joseph Doherty d84b066c62 [M3] mxaccess-galaxy: GalaxyTagMetadata + parser + Resolver trait + SQL 
Lands M3 stream A — the pure-Rust foundation of the Galaxy resolver:
the data type, the tag-reference parser, the async trait, and the
canonical SQL strings. Unblocks F13 (NmxClient::write_* wrappers depend
on GalaxyTagMetadata) without pulling in tiberius yet.

New
- metadata.rs (~195 LoC, 7 tests) — GalaxyTagMetadata record (port of
  cs:6-73). Includes is_buffer_property + to_reference_handle(galaxy_id)
  bridging into mxaccess-codec::MxReferenceHandle::from_names.
- parser.rs (~330 LoC, 12 tests) — ParsedTagReference parser. Handles
  Object.Attribute (1 candidate), Object.Primitive.Attribute (2
  candidates: primitive-attr first, dotted-attr second per cs:181-185),
  and the case-insensitive .property(buffer) suffix. Pure-Rust, no I/O.
- resolver.rs (~200 LoC, 5 tests including a tokio-driven InMemoryResolver
  proving the trait is implementable without SQL) — async Resolver trait
  + ResolverError. Default browse returns Backend("not implemented") so
  read-only backends don't need to override it.
- sql.rs (~280 LoC, 5 smoke tests) — RESOLVE_SQL + BROWSE_SQL constants
  ported byte-for-byte from cs:208-432. Available publicly so any
  backend (the planned tiberius impl, a wwtools/grdb snapshot replay,
  etc.) can grab the canonical query.

Cargo.toml: added mxaccess-codec (path), async-trait, thiserror;
tokio added as dev-dependency for the resolver-trait async tests.

Deliberately deferred to a later iteration:
- The tiberius-backed Resolver impl behind the galaxy-resolver feature.
- ToValueKind / TryGetValueKind / ProjectWriteValue helpers on
  GalaxyTagMetadata (cs:41-72) — these need a MxDataType -> MxValueKind
  lookup that the codec doesn't currently expose; landing them with
  F13's write-helper iteration keeps the iteration coherent.

Test count delta: 397 -> 427 (+30). All four DoD gates green.
Open followups touched: F13 prerequisite (GalaxyTagMetadata) now in
place; F13 itself stays open until the write helpers wire it up.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 08:17:16 -04:00
Joseph Doherty 0c772d273d [M3] mxaccess-nmx: NmxClient — 9 raw INmxService2 opnums (stream B) 
Lands M3 stream B raw opnum surface: an async NmxClient over the
mxaccess-rpc transport that dispatches all 9 INmxService2 procedures
(GetPartnerVersion, RegisterEngine2 + WithoutCallback, UnregisterEngine,
Connect, AddSubscriberEngine, RemoveSubscriberEngine,
SetHeartbeatSendInterval, TransferData) plus a NonZeroHresult error
variant that mirrors ThrowIfFailed (cs:563-574).

New
- crates/mxaccess-nmx/src/client.rs (~580 LoC, 8 tests including 5
  real-socket tokio tests against a hand-rolled DCE/RPC server) — port
  of the raw opnum surface from ManagedNmxService2Client.cs.
- NmxClient::connect builds the NTLM-packet-integrity bind path; for
  tests, NmxClient::from_bound_transport accepts a transport bound any
  way the caller likes (the test server doesn't validate signatures).
- fresh_orpc_this generates a per-call Cid via rand::random(), mirroring
  the .NET reference's Guid.NewGuid() at every call site.
- NmxClientError::NonZeroHresult unifies the .NET reference's
  Marshal.ThrowExceptionForHR + InvalidOperationException branches so
  callers see one typed surface for "transport-OK + LMX rejected".

Cargo.toml: added tokio, tracing, thiserror, rand to mxaccess-nmx.

Two layers of the .NET reference are deliberately out of scope this
iteration; both logged as new followups in design/followups.md:

- F12 (P1): the auto-resolving Create() factory, which needs windows-rs
  COM activation (gated by F6) + ComObjRefProvider port.
- F13 (P1): the high-level Write*/Advise*/UnAdvise/RegisterReference
  helpers, which depend on GalaxyTagMetadata from M3 stream A (the
  Galaxy SQL resolver crate, not yet started).

Test count delta: 389 -> 397 (+8). All four DoD gates green.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 08:06:15 -04:00
Joseph Doherty 432f1102b7 [M2/M3] mxaccess-rpc: tokio DCE/RPC TCP transport (DceRpcTcpClient port) 
Lands the async DCE/RPC TCP client — the transport that bridges the M2
PDU codec to a real socket. Unblocks M3 stream B (mxaccess-nmx, the
NmxClient) and brings F9 (ResolveOxid wrappers) within reach.

New
- transport.rs (~700 LoC, 10 tests including 2 real-socket tokio tests)
  — port of src/MxNativeClient/DceRpcTcpClient.cs.
  - DceRpcTcpClient::connect/bind/bind_with_managed_ntlm_packet_integrity/
    call/call_bound/call_bound_object — async over tokio::net::TcpStream.
  - encode_packet_integrity_request: 4-byte 0xBB pad + 8-byte AuthTrailer
    + 16-byte NtlmClientContext::sign signature, frag_length and
    auth_length rewritten in the embedded header per cs:201-250.
  - encode_request_bytes: PFC_OBJECT_UUID flag (0x80) and inserted
    16-byte object UUID slot per cs:269-278.
  - TransportError enum unifies io / codec / NTLM / fault / not-connected
    surfaces. Mirrors DceRpcFaultException as the typed Fault variant.
  - NTLM_AUTH_CONTEXT_ID = 79232 = 0x13580 (cs:90,133) exposed publicly.

Deliberately skipped: BindWithNtlmConnect / BindWithNtlmPacketIntegrity
(SSPI flavours at cs:55-63,108-149) — those wrap .NET's
System.Net.Security.SspiClientContext, which has no portable analogue.
Managed-NTLM path covers what the production Rust client needs.

mxaccess-rpc/Cargo.toml: added tokio (workspace-pinned).

design/followups.md: F9 downgraded P1 → P2 (transport landed; only the
two pure-codec ResolveOxid wrappers remain).

Test count delta: 354 -> 364 (+10).
Open followups touched: F9 partially advanced.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 07:47:42 -04:00
Joseph Doherty b0954b2672 [M2] mxaccess-callback: tokio TCP exporter (wave 3 main) 
Lands the M2 wave 3 main course — the INmxSvcCallback callback exporter.
Pure-tokio TCP server that mirrors src/MxNativeClient/ManagedCallbackExporter.cs
and lets a Rust client receive callbacks from NmxSvc.exe.

New
- exporter.rs (~700 LoC, 10 tests) — port of ManagedCallbackExporter.cs.
  CallbackExporter::bind starts a TcpListener + accept loop; per-connection
  serve task walks Bind / AlterContext / Request / Auth3 PDUs and dispatches
  IRemUnknown (opnums 3/4/5) and INmxSvcCallback (opnums 3/4) requests.
  Hand-rolled BindAck encoder mirroring cs:226-254 (single acceptance entry,
  NDR20 transfer syntax).
- ExporterIdentities { oxid, oid, callback_ipid, rem_unknown_ipid } — exposes
  both `random()` (production) and `fixed()` (tests). Mirrors the .NET
  RandomUInt64 + Guid.NewGuid pattern at cs:14-20.
- CallbackEvent enum — typed diagnostic stream replacing .NET's
  List<string> log (cs:12,33-42,315-321). Variants: ClientConnected,
  AcceptError, Bind, Auth3Ignored, Request, RemQueryInterface,
  CallbackInvoked, UnhandledRequest, ClientDisconnected, ProtocolError.
- IUNKNOWN_IID const re-exported alongside the other IIDs.

Tests cover real-socket round-trips: Bind+RemQueryInterface (with IUNKNOWN
returning S_OK), Bind+unknown opnum -> Fault, Bind+DataReceived ->
CallbackInvoked event + 12-byte success response, and graceful shutdown.

Test count delta: 344 -> 354 (+10).
Open followups touched: none new. F2 (verify_signature path) still
gated on a live status-frame fixture under tests/fixtures/m2-status-frame/.
F6 / F9 still need the windows-rs and DceRpcTcpClient ports respectively.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 07:36:41 -04:00
Joseph Doherty 95bd218183 [M2] mxaccess-rpc: NTLMv2 + DCE/RPC PDU + OBJREF parser (wave 1) 
Lands M2 wave 1 — three pure-Rust modules under crates/mxaccess-rpc with
60 unit tests. Each is a 1:1 port of one .NET reference file:

- ntlm.rs (1137 LoC, 19 tests) — `ManagedNtlmClientContext.cs`. NTLMv2
  challenge/response, Type1/Type3 builders, sign() with RC4-sealed checksum
  and per-call sequence advance. Manual `Debug` impl that hides credentials;
  not Clone (rc4 0.2 cipher state is non-Clone). Pure-Rust crypto via
  hmac/md-5/md4/rc4 v0.2/rand v0.8 (rc4 0.2 chosen per design/review.md:78).
- pdu.rs (1573 LoC, 33 tests) — `DceRpcPdu.cs` + auth-trailer types from
  `DceRpcAuthentication.cs`. Bind/AlterContext/Auth3/Request/Response/Fault
  PDUs, NDR20 transfer syntax, auth_value with 4-byte alignment padding,
  preserved-byte fields per CLAUDE.md unknown-bytes rule.
- objref.rs (~470 LoC, 11 tests including a 366-byte captured OBJREF
  round-trip) — `ComObjRef.cs`. MEOW signature, OXID/OID/IPID, dual-string
  array with printable-ASCII escaping and security-binding boundary.
  ComObjRefProvider.cs deferred (windows-rs Win32 wrapper — see F6).

Every wire-byte claim cites src/MxNativeClient/<file>.cs:LINE per
CLAUDE.md "no fabricated protocol behaviour" rule.

Test count delta: 217 → 277 (+60)
Open followups touched: F1–F8 (new — see design/followups.md)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 06:54:39 -04:00
Joseph Doherty fe2a6db786 Initial project state: .NET reference, design, Rust port (M0+M1), evidence
rust / build / test / clippy / fmt (push) Has been cancelled
Details 
Layout:
- src/                    .NET 10 x64 reference: MxNativeCodec, MxNativeClient,
                          MxAsbClient, probes, tests, harnesses. Executable spec.
- design/                 Architectural plan for the Rust port (M0–M6), error
                          model, protocol invariants, risks (R1–R16), adversarial
                          review log (review.md).
- rust/                   Rust workspace. M0 skeleton + M1 codec parity.
                          mxaccess-codec: 215 unit tests + 2 cross-implementation
                          parity tests (byte-identical against .NET reference).
                          Other crates are M0 stubs awaiting M2+.
- captures/               Frida + netsh + pcap evidence per CLAUDE.md
                          ("captures are evidence, not throwaway logs").
- analysis/               Decompiled C# (frida/proxy/decompiled-*),
                          Ghidra exports for native DLLs (`exports/` only —
                          working state at `projects/` and AVEVA's input
                          binaries at `input/` are gitignored).
- docs/                   Reverse-engineering reference docs.
- tools/                  Setup-LiveProbeEnv.ps1 (Infisical credential fetcher),
                          Compute-Crc.ps1 (.NET parity helper).
- .github/workflows/      Rust CI: fmt + build + test + clippy on Windows.
- LICENSE                 MIT (Joseph Doherty, 2026).

Verified:
- cargo test --workspace → 217 passed (215 unit + 2 .NET parity), 0 failed
- cargo clippy --workspace -- -D warnings → clean
- cargo fmt --all -- --check → clean
- cargo publish --dry-run -p mxaccess-codec → packages cleanly

Excluded from history (see .gitignore):
- **/bin, **/obj, **/target — build artifacts
- analysis/ghidra/projects/ — Ghidra working state (regenerable)
- analysis/ghidra/input/ — AVEVA proprietary DLLs (vendor IP)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 06:21:00 -04:00