Commit 0702551 regenerated all three protos with grpcio-tools 1.81.1 /
protobuf 6.33.5, stamping every _pb2_grpc.py with GRPC_GENERATED_VERSION
'1.81.1' and every _pb2.py with 'Protobuf Python Version: 6.33.5'. The
project pins grpcio>=1.80,<2 and the installed runtime is 1.80.0, so the
1.81.1 stamp caused a version error at import time, breaking pytest.
Fix: restore all generated files to the main baseline (grpcio 1.80.0,
protobuf 6.31.1 stamps), then regenerate only mxaccess_gateway.proto
using an isolated venv pinned to grpcio==1.80.0, grpcio-tools==1.80.0,
protobuf==6.31.1. The net diff vs main is a single-file change to
mxaccess_gateway_pb2.py whose DESCRIPTOR blob now encodes the new
MxSparseArray / MxSparseElement messages and sparse_array_value field.
No _pb2_grpc.py and no galaxy/worker _pb2.py files change.
Fix the MxSparseArray type-defaults table in gateway.md to use the real MxDataType
enum member names (Integer, not Integer/LongInteger; Time, not Time/Timestamp) and
clarify the int64 note. Add a docstring sentence to write_array_elements in the
Python client noting that indices and total_length must be non-negative.
Handles the new MxSparseArray wire type (proto field 19 on MxValue::Kind):
- value.rs: map SparseArrayValue to MxValueProjection::Unset (write-only; never emitted on read path)
- session.rs: add write_array_elements() that builds the sparse proto value and delegates to write()
- tests: three unit tests asserting proto shape, empty-elements case, and read-path Unset projection
- README: document write_array_elements default-fill semantics and bare-name [] normalisation
Regenerate Go proto types from mxaccess_gateway.proto so MxSparseArray,
MxSparseElement, and MxValue_SparseArrayValue appear in the generated
package; add MxSparseArray/MxSparseElement type aliases to types.go;
add Session.WriteArrayElements and the unexported buildSparseArrayValue
builder; add three unit tests covering the sparse oneof structure,
empty-map case, and the round-trip through WriteArrayElements; update
README with default-fill reset semantics and auto-normalize note.
Regenerate Python proto bindings to pick up MxSparseArray/MxSparseElement/
sparse_array_value from the shared mxaccess_gateway.proto. Add
Session.write_array_elements which builds an MxValue(sparse_array_value=…)
from a {index→scalar} dict and delegates to the existing write(). Add 8 pytest
tests covering builder correctness and full round-trip wire shape. Update
README with a default-fill semantics paragraph and bare-name array-write note.
Adds a public WriteArrayElementsAsync helper on MxGatewaySession that builds
an MxValue{SparseArrayValue} and delegates to the existing WriteAsync. Extracts
the MxValue construction into an internal static BuildSparseArray builder for
unit-testability. Two new tests cover builder output shape and the full write
command path. README documents the reset (not preserve) semantics alongside
the existing whole-array guidance.
A consuming project hit two MXAccess parity surprises: a plain Write only
records its user_id when the item has an active supervisory advise (the path
to take when not authenticating), and array writes replace the whole array
rather than patching individual elements. Document both across the five client
READMEs and gateway.md's compatibility baseline, and expose the missing
advise-supervisory subcommand in the go/python/rust/java CLIs (plus the .NET
help text) so callers can establish the supervisory advise without dropping to
the raw command API.
Same \u-in-comment issue as the production file (MxGatewayCliTests.java:69).
Reworded to plain ASCII. Scanned all agent-touched Java files; no other stray
unicode-escape sequences remain.
The Client.Java-041 escaping fix left a literal \u00XX in a // comment; Java
processes \u unicode escapes even in comments, so it failed to compile on
windev (MxGatewayCli.java:2223 illegal unicode escape). Reworded the comment to
plain ASCII. The escaping code itself (String.format with the doubled-backslash
format string) was already correct.
Client.Java-040..048, Worker.Tests-034/035/036. Edits applied on the Mac,
which has no JRE and cannot build the x86+MXAccess worker tests; findings are
marked In Progress pending gradle + x86 build verification on windev. Do not
mark Resolved until verified there.
Server-054/055/056, Contracts-020/021/022, Tests-036/038/039,
IntegrationTests-030/031/032 (+033 deferred to live rig),
Client.Dotnet-026/028/029 (+027 won't-fix), Client.Go-030..034,
Client.Python-032..036, Client.Rust-033..038.
Key fix: SessionEventDistributor orphaned a subscriber that registered after
the pump completed but before disposal (Server-056) -> register paths now
complete late registrants under _lifecycleLock; regression test added. The
racy dashboard-mirror gRPC test made deterministic (Tests-039).
Verified green locally: gateway Tests targeted classes (GatewaySession,
SessionEventDistributor, GatewayOptionsValidator, ProtobufContractRoundTrip,
GatewaySessionDashboardMirror) + dotnet/go/python/rust client suites.
Add --parent-gobject-id (integer) to the galaxy-browse CLI command so the
Python client matches the Go (-parent) and Rust (--parent-gobject-id) CLIs.
When set, drives BrowseChildren paging via browse_children_raw (page size 500,
repeated-token guard) and renders the same JSON node shape (flattened object
fields + hasChildrenHint + empty children array) and indented-text tree as the
root-walk path. --depth is ignored on the parent path with a one-line stderr
warning, matching the Go/Rust behaviour. Tests added in TDD order.
Replaces the raw IllegalArgumentException thrown by GalaxyBrowseCommand for
--depth < 0 with a CommandLine.ParameterException so picocli surfaces a clean
single-line error instead of an unhandled stack trace. Adds an upper bound of
50 (matching the Python client) so --depth > 50 is also rejected cleanly.
Emits a stderr warning when --parent 0 is supplied explicitly, matching
Go/Rust client behaviour, because gobject id 0 is the server's root-walk
sentinel and passing it via --parent is almost always a mistake.
Adds three new tests: negative depth, depth > 50, and the --parent 0 warning path.
- D4: add 'ping' subcommand (MX_COMMAND_KIND_PING / PingCommand{message}),
accepting --session-id and optional --message (default "ping"); prints the
worker's echoed diagnostic message.
- D8-java: add 'galaxy-browse' subcommand over browse()/LazyBrowseNode.expand()
and raw BrowseChildren paging for --parent. JSON node shape matches the
cross-client surface (flattened object fields + hasChildrenHint + nested
children array).
- D9-java: make galaxy-test-connection / galaxy-last-deploy the primary names,
keeping galaxy-test / galaxy-deploy-time as deprecated picocli aliases.
- Tests for ping, galaxy-browse JSON hasChildrenHint key, and alias resolution.
- README updated for the new/renamed subcommands.
Rename the browse JSON key from hasChildren to hasChildrenHint to match the
Rust and Python CLIs and the library property name (HasChildrenHint). Update
the text-output label to match. Add a one-line stderr warning when -parent 0
is supplied, since 0 is the server root sentinel and omitting -parent is the
intended way to walk from the root.
Rename the anonymous-object member `hasChildren` → `hasChildrenHint` so the
serialized JSON key matches the Rust and Python CLIs and the library property
HasChildrenHint. Also update the text-output suffix to `hasChildrenHint=` for
consistency.
Guard _galaxy_browse against unbounded recursion by rejecting --depth
values outside [0, 50] with a descriptive BadParameter. Add test coverage
for --depth 99 and --depth -1 rejection, and assert _text is never
present in the JSON output from galaxy-browse.
Warn on stderr when --parent-gobject-id and --depth>0 are both supplied
since depth is silently ignored in the single-level parent path. Also
updates the --depth arg doc to document this. Extracts BROWSE_PAGE_SIZE
const (500) with a cross-reference to galaxy.rs instead of a bare literal.
Removes three redundant .clone() calls in BrowseChildrenOptions construction
since the originals are not used after the struct is built.
TestRunPingJSON now verifies the fake gateway's echoed text appears in
the serialised reply body, catching any future wiring regression that
maps PingRaw to the wrong proto field. runPing gains a one-line comment
explaining why DiagnosticMessage carries the echo, why the kind-string
fallback exists, and why writeCommandOutput is not reused on the
plain-text path.
Add PingRaw to Session (session.go), runPing to the CLI dispatch
(main.go), and three tests covering plain-text echo, JSON output,
and missing-session-id validation (main_test.go). Default message
is "ping"; gateway echo is read from DiagnosticMessage, falling
back to the kind string if absent.
Add Write2/Write2Raw to the Go client Session, mirroring the existing
Write/WriteRaw pair, so all five language clients now expose write2.
Includes three TDD tests covering payload propagation, raw-reply return,
and nil-value rejection.
Extract the TLS-without-CA guard into a shared `build_tls_config` helper
in options.rs so both GatewayClient and GalaxyClient use identical logic.
GalaxyClient previously had no guard, so TLS-without-CA produced a cryptic
tonic handshake failure; it now returns the same actionable InvalidEndpoint
error. The guard message notes that a server-name override affects SNI but
does not pin trust. Add --require-certificate-validation to ConnectionArgs
in the CLI binary. Add a mirror test for GalaxyClient in tests/tls.rs.
Mirror MxGatewayClient's three-branch handler structure in GalaxyRepositoryClient
(CA-pin / lenient accept-all / OS trust) so the Galaxy endpoint works against the
gateway's self-signed cert under the default lenient posture. Expose an internal
CreateHttpHandlerForTests seam for unit testing. Add RemoteCertificateNameMismatch
rejection at the top of both CA-pinned callbacks so a pinned-CA connection truly
verifies the host. Strengthen existing lenient test to invoke the callback and assert
it returns true; add mirrored Galaxy-client handler tests.
Joseph Doherty