FOCAS PR 1 � Scaffolding + Core #124

dohertj2 · 2026-04-19T19:49:44-04:00

dohertj2 commented

2026-04-19 19:49:44 -04:00

Driver.FOCAS project with FocasHostAddress (focas://host:port), FocasAddress (PMC X/Y/F/G/R/D/C/K/A/E/T + byte + optional bit, PARAM:n/bit, MACRO:n), 7 atomic data types, FWLIB EW_* status mapper, IFocasClient abstraction with UnimplementedFocasClientFactory default. 65 new tests. Merges to v2.

dohertj2 added 1 commit 2026-04-19 19:49:46 -04:00

FOCAS PR 1 — Scaffolding + Core (FocasDriver skeleton + address parser + stub client). New Driver.FOCAS project for Fanuc CNC controllers (FS 0i/16i/18i/21i/30i/31i/32i/Series 35i/Power Mate i) talking via the Fanuc FOCAS/2 protocol. No NuGet reference to a FOCAS library — FWLIB (Fwlib32.dll) is Fanuc-proprietary + per-customer licensed + cannot be legally redistributed, so the driver is designed from the start to accept an IFocasClient supplied by the deployment side. Default IFocasClientFactory is UnimplementedFocasClientFactory which throws with a clear deployment-docs pointer at Create time so misconfigured servers fail fast rather than mysteriously hanging. Matches the pattern other drivers use for swappable wire layers (Modbus IModbusTransport, AbCip IAbCipTagFactory, TwinCAT ITwinCATClientFactory) — but uniquely, FOCAS ships without a production factory because of licensing. FocasHostAddress parses focas://{host}[:{port}] canonical form with default port 8193 (Fanuc-reserved FOCAS Ethernet port). Default-port stripping on ToString for roundtrip stability. Case-insensitive scheme. Rejects wrong scheme, empty body, invalid port, non-numeric port. FocasAddress handles the three addressing spaces a FOCAS driver touches — PMC (letter + byte + optional bit, X/Y for IO, F/G for PMC-CNC signals, R for internal relay, D for data table, C for counter, K for keep relay, A for message display, E for extended relay, T for timer, with .N bit syntax 0-7), CNC parameters (PARAM:n for a parameter number, PARAM:n/N for bit 0-31 of a parameter), macro variables (MACRO:n). Rejects unknown PMC letters, negative numbers, out-of-range bits (PMC 0-7, parameter 0-31), non-numeric fragments. FocasDataType — Bit / Byte / Int16 / Int32 / Float32 / Float64 / String covering the atomic types PMC reads + CNC parameters + macro variables return. ToDriverDataType widens to the Int32/Float32/Float64/Boolean/String surface. FocasStatusMapper covers the FWLIB EW_* return-code family documented in the FOCAS/1 + FOCAS/2 references — EW_OK=0, EW_FUNC=1 → BadNotSupported, EW_OVRFLOW=2/EW_NUMBER=3/EW_LENGTH=4 → BadOutOfRange, EW_PROT=5/EW_PASSWD=11 → BadNotWritable, EW_NOOPT=6/EW_VERSION=-9 → BadNotSupported, EW_ATTRIB=7 → BadTypeMismatch, EW_DATA=8 → BadNodeIdUnknown, EW_PARITY=9 → BadCommunicationError, EW_BUSY=-1 → BadDeviceFailure, EW_HANDLE=-8 → BadInternalError, EW_UNEXP=-10/EW_SOCKET=-16 → BadCommunicationError. IFocasClient + IFocasClientFactory abstraction — ConnectAsync, IsConnected, ReadAsync returning (value, status) tuple, WriteAsync returning status, ProbeAsync for IHostConnectivityProbe. Deployment supplies the real factory; driver assembly stays licence-clean. FocasDriverOptions + FocasDeviceOptions + FocasTagDefinition + FocasProbeOptions — one instance supports N CNCs, tags cross-key by HostAddress + use canonical FocasAddress strings. FocasDriver implements IDriver only (PRs 2-3 add read/write/discover/subscribe/probe/resolver). InitializeAsync parses each device HostAddress + fails fast on malformed strings → Faulted health. 65 new unit tests in FocasScaffoldingTests covering — 5 valid host forms + 8 invalid + default-port-strip ToString, 12 valid PMC addresses across all 11 canonical letters + 3 parameter forms with + without bit + 2 macro forms, 10 invalid address shapes, canonical roundtrip theory, data-type mapping theory, FWLIB EW_* status mapping theory (9 codes + unknown → generic), DriverType, multi-device Initialize + address parsing, malformed-address fault, shutdown, default factory throws NotSupportedException with deployment pointer + Fwlib32.dll mention. Total project count 31 src + 20 tests; full solution builds 0 errors. Other drivers untouched. 285799a954

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

dohertj2 merged commit c13fe8f587 into v2

2026-04-19 19:49:49 -04:00

dohertj2 referenced this issue from a commit

2026-04-19 19:49:50 -04:00

Merge pull request (#124) - FOCAS scaffolding

dohertj2 referenced this issue from a commit

2026-04-19 23:04:15 -04:00

ExternalIdReservation merge inside FinaliseBatchAsync. Closes task #197. The FinaliseBatch docstring called this out as a narrower follow-up pending a concurrent-insert test matrix, and the CSV import UI PR (#163) noted that operators would see raw DbUpdate UNIQUE-constraint messages on ZTag/SAPID collision until this landed. Now every finalised-batch row reserves ZTag + SAPID in the same EF transaction as the Equipment inserts, so either both commit atomically or neither does. New MergeReservation helper handles the four outcomes per (Kind, Value) pair: (1) value empty/whitespace → skip the reservation entirely (operator left the optional identifier blank); (2) active reservation exists for same EquipmentUuid → bump LastPublishedAt + reuse (re-finalising a batch against the same equipment must be idempotent, e.g. a retry after a transient DB blip); (3) active reservation exists for a DIFFERENT EquipmentUuid → throw ExternalIdReservationConflictException with the conflicting UUID + originating cluster + first-published timestamp so operator sees exactly who owns the value + where to resolve it (release via sp_ReleaseExternalIdReservation or pick a new ZTag); (4) no active reservation → create a fresh row with FirstPublishedBy = batch.CreatedBy + FirstPublishedAt = transaction time. Pre-commit overlap scan uses one round-trip (WHERE Kind+Value IN the batch's distinct sets, filtered to ReleasedAt IS NULL so explicitly-released values can be re-issued per decision #124) + caches the results in a Dictionary keyed on (Kind, value.ToLowerInvariant()) for O(1) lookup during the row loop. Race-safety catch: if another finalise commits between our cache-load + our SaveChanges, SQL Server surfaces a 2601/2627 unique-index violation against UX_ExternalIdReservation_KindValue_Active — IsReservationUniquenessViolation walks the inner-exception chain for that specific signature + rethrows as ExternalIdReservationConflictException so the UI shows a clean message instead of a raw DbUpdateException. The index-name match means unrelated filtered-unique violations (future indices) don't get mis-classified. Test-fixture Row() helper updated to generate unique SAPID per row (sap-{ZTag}) — the prior shared SAPID="sap" worked only because reservations didn't exist; two rows sharing a SAPID under different EquipmentUuids now collide as intended by decision #124's fleet-wide uniqueness rule. Four new tests: (a) finalise creates both ZTag + SAPID reservations with expected Kind + Value; (b) re-finalising same EquipmentUuid's ZTag from a different batch does not create a duplicate (LastPublishedAt refresh only); (c) different EquipmentUuid claiming the same ZTag throws ExternalIdReservationConflictException with the ZTag value in the message + Equipment row for the second batch is NOT inserted (transaction rolled back cleanly); (d) row with empty ZTag + empty SAPID skips reservation entirely. Full Admin.Tests suite 85/85 passing (was 81 before this PR, +4). Admin project builds 0 errors. Note: the InMemory EF provider doesn't enforce filtered-unique indices, so the IsReservationUniquenessViolation catch is exercised only in the SQL Server integration path — the in-memory tests cover the cache-level conflict detection in MergeReservation instead, which is the first line of defence + catches the same-batch + published-vs-staged cases. The DbUpdate catch protects only the last-second race where two concurrent transactions both passed the cache check.

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: dohertj2/lmxopcua#124