* Fix logger concurrency correctness and TRX write reliability

Both TrxLogger and HtmlLogger register event handlers that are invoked
concurrently during parallel test execution, but several pieces of their
mutable state were plain non-thread-safe collections.

TrxLogger:
- _runLevelErrorsAndWarnings is now a ConcurrentQueue<RunInfo>.
- _runLevelStdOut is now a ConcurrentQueue<string> instead of a StringBuilder;
  the run level informational message is materialized on read with the same
  AppendLine-per-message formatting as before.
- CreateTestRun had a check-then-act race that could create more than one
  TestRun. It is replaced by GetOrCreateTestRun, which uses double checked
  locking, and LoggerTestRun is published with Volatile so that a reader that
  sees a non-null reference also sees a fully initialized TestRun.
- TestElementAggregation._testLinks was a plain Dictionary mutated through a
  non-atomic ContainsKey + Add. Mutation and enumeration are now guarded by a
  lock. A lock is used rather than ConcurrentDictionary so that the insertion
  order of test links, which is meaningful for ordered tests, is preserved.
- PopulateTrxFile only caught UnauthorizedAccessException, so a disk full or
  any other IO failure silently dropped the TRX file with an unhandled
  exception. It now reports IOException, UnauthorizedAccessException,
  XmlException, NotSupportedException and SecurityException using the new
  localized TrxLoggerWriteFailed resource.

HtmlLogger:
- TestResultHandler used TryGetValue + TryAdd + ResultCollectionList.Add, which
  could publish duplicate collections for the same source. It now uses the
  atomic GetOrAdd and only the winning thread publishes the collection.
- ResultList, FailedResultList, InnerTestResults and the run level message
  lists are appended through lock protected helpers, so no result or message
  can be lost. The lists are still created lazily, so the null versus empty
  semantics observed by the serialized XML are unchanged.

Fixes #16320

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 32aa1e90-526a-4b49-97cc-1297638d7520

* Correct two inaccurate code comments from review feedback

- TestRunDetails._syncLock no longer describes ResultCollectionList as
  lazily created, since that field is initialized inline. Only the run
  level message lists are created on first use.
- GetOrCreateTestRun no longer says Started is set to DateTime.Now in
  Initialize; TestRunStartTime is captured with DateTime.UtcNow.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 32aa1e90-526a-4b49-97cc-1297638d7520

---------

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 32aa1e90-526a-4b49-97cc-1297638d7520

* Restore custom data collector platform tests

Use the existing out-of-process collector test asset through the supported discovery path, and complete its secondary event channel before ending each run.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

🤖

* Copy collector from resolved project output

Use the ProjectReference target path instead of assuming its configuration matches the solution configuration.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

🤖

* Always close collector event client

Reject invalid event ports and stop the secondary communication client even when setup or validation fails.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

Copilot-Session: 74aae65f-0435-4727-b13a-ff1e80980b3f

🤖

* Revive socket client and server tests

Stop socket listeners deterministically and serialize socket server startup, accept, and shutdown so port 0 listeners cannot escape Stop.

Replace unbounded waits and stale post-close assertions, and re-enable the two stable socket test classes.

🤖

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Fix SocketServer accept lock scope

🤖

* Harden SocketServer shutdown regression

🤖

* Scope SocketServer shutdown handling

🤖

---------

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

…ient.Sources (#16300)

* Retarget the MTP client onto Microsoft.Testing.Platform.ServerClient.Source

testfx now ships vstest's MTP server-mode JSON-RPC client as a source-only
package built from the MTP server's own protocol and serialization source, so
the wire format cannot drift from the server.

Delete vstest's transport core (MtpServerConnection, MtpJson, MtpConstants,
MtpClientHelpers) and retarget the glue onto the package's IMtpServerClient:
launch via MtpServerClient.Launch, drive Initialize/Discover/Run/Exit, read
node updates from the TestNodesUpdated event with typed MtpTestNodeUpdate
accessors, and bridge EqtTrace through DelegateMtpClientLogger.
MtpClientOptionsFactory centralizes option construction and log-level mapping.

The package is a compile-time source dependency (PrivateAssets=all), so no
runtime dependency and no public API are added.

Blocked on testfx publishing the package (microsoft/testfx#10085); references
an interim local feed, so CI cannot restore it yet.

* Commit the interim local MTP client feed so restore works everywhere

NuGet.config pointed local-mtp at the absolute path Q:\q\local-mtp-feed, which
is machine-local and does not exist in CI, so restore failed with an incorrect
path. Move the feed under the repo at eng/local-mtp-feed, point NuGet.config at
that repo-relative path, and commit the package into the feed. .gitignore keeps
ignoring *.nupkg but adds a negation for eng/local-mtp-feed/*.nupkg so the feed
package is tracked.

The package is the fresh Design-A drop of
Microsoft.Testing.Platform.ServerClient.Source 2.4.0-dev, which builds
CrossPlatEngine clean on net462, netstandard2.0, and net8.0 (0 errors, 0
warnings) with the retargeted glue. Interim only; remove the feed once
microsoft/testfx#10085 ships the package to a public feed.

🤖

* Order remote NuGet feeds before the interim local feed in test asset restore

The acceptance tests restore the TestAssets solution, which transitively
restores product projects like CrossPlatEngine that now reference the interim
local-mtp feed. Passing that local-folder feed to dotnet restore alongside the
remote https feeds triggered two NuGet quirks, both surfacing as NU1301: a
relative --source path is rooted at each restored project's directory, and a
local-folder source placed before the remote sources mis-normalizes the https
URLs into per-project relative paths.

Resolve relative local-folder sources to absolute paths and emit the remote
sources first so all local-folder sources come last; remote feeds keep their
configured order. Only needed while the MTP client package lives on the interim
local feed, and harmless once testfx#10085 ships it to a public feed.

🤖

* Key MTP environment variable dictionary case-insensitively on Windows

Both places that collect environment variables for the MTP application
launch now share one comparer: case-insensitive on Windows, case-sensitive
elsewhere. Before, the runsettings path used that comparer but the
data-collector-only path used a plain ordinal dictionary, so a run with no
runsettings variables but with data-collector variables lost the
case-folding the classic testhost path applied on Windows. The package
options dictionary is ordinal, so deduping here preserves the classic
Windows semantics before the values reach it.

🤖

* Consume official B-fixed MTP client source drop (testfx#10085)

Replaces the interim 2.4.0-dev pack with the official drop that fixes the
STJ number-decode bug: untyped JSON numbers were hard-cast to Int32, so node
bags carrying doubles (durations) or longs (timestamps) threw FormatException
and faulted the MTP read loop on the net8 client. The fix decodes numbers
generically (ReadNumber: TryGetInt32 -> TryGetInt64 -> TryGetUInt64 -> double).

Pinned to the unique version 2.4.0-dev.20260721161520 to avoid NuGet
same-version cache collisions while the package is served from the committed
local feed.

MtpUnderVstestTests: net11.0 (STJ) axis now 7/7 (was 0/7); net481 (Jsonite)
axis 5/7. The 2 remaining failures are a pre-existing net462 TRX-logger load
issue that also breaks classic non-MTP trx tests, unrelated to this retarget.

🤖

* Align interim MTP client pin to the coordinator's canonical numberfix drop

Swaps the interim feed pack and pin from the timestamped unique
2.4.0-dev.20260721161520 to the coordinator's canonical uniquely-named drop
2.4.0-dev.numberfix (MD5 FC7F7A9F68EF482718B61DC9DA5F38B4). Byte-equivalent
fixed content -- the packed net8 Json.Deserializers.cs decodes untyped JSON
numbers via ReadNumber at both sinks (L55/L97, helper L344), same as the prior
drop -- this only adopts the stable canonical interim identity the package
owner is standardizing on across consumers.

Validation unchanged: MtpUnderVstestTests net11.0 (STJ) axis 7/7, full suite
12/14 (the 2 remaining failures are the pre-existing net462 TRX-logger load
issue, unrelated to this retarget).

🤖

* Add MTP converter/options unit tests and fix numeric and trait coercion

The retarget onto Microsoft.Testing.Platform.ServerClient.Source left the MTP
glue with no unit coverage at all - the only tests were the end-to-end
MtpUnderVstestTests. The conversion code is now pure and dependency-free, so
cover it directly.

Add MtpTestNodeConverterTests and MtpClientOptionsFactoryTests (55 tests)
covering the normalized-Node contract, per-formatter number boxing, outcome
mapping, the action-node filter, vstest bridge properties, standard
output/error, traits, duration and log-level mapping.

Three fixes fall out of writing them:

- TryGetRawInt wrapped out-of-range values with unchecked((int)l), turning a
  bad line number into a plausible-looking wrong answer. Range-check instead so
  the property stays at its visibly-unset default.
- AddTraits collapsed every non-string trait value to an empty string. The two
  formatters box JSON scalars differently, so a numeric or boolean trait was
  silently dropped on one formatter and kept on the other. Format invariantly.
- MtpClientOptionsFactory re-read VSTEST_CONNECTION_TIMEOUT and hardcoded the
  90-second default instead of calling EnvironmentHelper.GetConnectionTimeout,
  which seven other vstest call sites already use and which also traces the
  override.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: cd0793f4-d530-44b6-b881-fed3be6aa52f

* Fix MTP client shutdown and fail loudly on a missing node uid

Retargeting onto the source package changed exit from a fire-and-forget
notification into an awaited request/response call, which introduced two
regressions:

- Exit was awaited on the run's own cancellation token. Cancelling or aborting
  a run is exactly when that token is already cancelled, so ExitAsync threw
  immediately and the graceful shutdown handshake was skipped in the one case
  it matters most.
- The await was unbounded, so a test application that never acknowledges exit
  would hang discovery or execution indefinitely. The notification it replaced
  could not block at all.

Route both proxy managers through MtpServerClientFactory: TryExit runs on its
own bounded token, swallows failures (the caller disposes the client next,
which tears the process down regardless), and is called from a finally block so
a failed or cancelled run still shuts the application down.

The factory also exposes a replaceable Launch delegate so the managers can be
driven against a fake server in unit tests; production always uses
MtpServerClient.Launch.

Separately, BuildUids substituted FullyQualifiedName when a TestCase carried no
MTP.TestNode.Uid. The server projects node.Uid alone when building a run filter
and never reads any other field, so that substitution produced a filter
matching nothing: the run reported success having executed zero of the tests
the user selected, with no error anywhere. Throw instead, with a comment
explaining why no fallback is correct.

Adds 15 tests covering the shutdown paths, the uid filter, and both manager
flows against a fake MTP server.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: cd0793f4-d530-44b6-b881-fed3be6aa52f

* Add non-ASCII MTP acceptance coverage for UTF-8 frame length

MTP frames declare Content-Length in UTF-8 bytes, but the transport shipped by
Microsoft.Testing.Platform.ServerClient.Source reads that number of characters:
it rents a char buffer of Content-Length and calls StreamReader.ReadBlockAsync.
For any frame carrying multi-byte UTF-8 the two disagree, so the reader
under-reads and leaves the body's tail to be parsed as the next frame's headers
- the connection desynchronizes from the following message onward.

vstest's deleted MtpServerConnection was byte-correct here (it read Content-Length
bytes into a byte[] and then UTF-8-decoded), so the retarget is a regression, not
an inherited defect. Client-to-server traffic is ASCII in practice, which is why
it has not surfaced; node updates flow the other way and carry user-authored test
names.

Give MtpMSTestProject a test whose display name mixes German umlauts (2 bytes
each), Japanese (3 bytes each) and an emoji (4 bytes, 2 chars), and mirror it in
MtpPureProject. Because the corruption lands on the message *after* the offending
one, its mere presence makes the whole run fail rather than just that test, so
every existing MTP scenario now exercises the transport with multi-byte content.
Adds a dedicated test asserting the name survives into the TRX.

These fail until the fix lands upstream in testfx.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: cd0793f4-d530-44b6-b881-fed3be6aa52f

* Narrow the non-ASCII MTP test name to the BMP and fix the collector count

Running the acceptance test revealed two things worth recording.

First, an end-to-end MTP run cannot reproduce the Content-Length byte-vs-char
framing bug: the .NET MTP server serializes with System.Text.Json, whose default
encoder escapes every non-ASCII character to \\uXXXX, so the bytes on the wire
are ASCII and the byte count coincidentally equals the character count. The
framing bug is real but has to be proved at the unit level against the transport
directly, which is what the companion testfx change does. This test is therefore
a name-integrity guard, and its comments now say so rather than overclaiming.

Second, the emoji originally in the name exposed a separate defect: astral-plane
characters are escaped by System.Text.Json as a surrogate pair and arrive in the
TRX as the literal text \\ud83c\\udf89 instead of the character. BMP characters
decode correctly. That is its own bug, tracked separately, so the name is
narrowed to BMP multi-byte characters (umlauts 2 bytes, Japanese 3 bytes) which
still exercise the byte-denominated length without tripping over it.

Also updates the out-of-proc data collector's expected per-test-case attachment
count, which follows the test count.

MtpUnderVstestTests: 16/16 on both console axes.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: cd0793f4-d530-44b6-b881-fed3be6aa52f

* Consume the MTP client drop with the Content-Length framing fix

Replaces the interim local-feed pack with a build of microsoft/testfx#10297,
which stacks the Content-Length byte/char fix onto #10085. The transport now
reads exactly Content-Length bytes and UTF-8-decodes them, symmetric with the
write path, and reads the headers through the same byte-level buffer so no
StreamReader can buffer part of the body across the boundary.

That drop also carries #10085's ServerRequestHandler signature change (the
result is now constrained to a serializable dictionary), so FakeMtpServerClient
is updated to match.

Verification on this drop:
- CrossPlatEngine builds clean on net462, netstandard2.0 and net8.0.
- MTP unit tests 140/140 (70 per axis, net11.0 and net481).
- MtpUnderVstestTests 16/16 on both console axes.

Note the 16/16: the two /logger:trx failures reported against the earlier drop
do not reproduce here, so they look like a local deployment issue rather than
anything in the retarget.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: cd0793f4-d530-44b6-b881-fed3be6aa52f

* Repin the interim MTP client to the uniquely-named utf8fix1 drop

Swaps the interim feed pack from the plain 2.4.0-dev build output to the
coordinator's canonical 2.4.0-dev.utf8fix1 drop of microsoft/testfx#10297.

Byte-equivalent content: all 184 contentFiles are identical between the two
packs, including TcpMessageHandler.cs with both ReadExactlyAsync and the
TrimPreamble BOM tolerance. Only the version metadata differs.

The rename is the point. While the package is served from a committed local
folder, NuGet caches by version, so a plain 2.4.0-dev risks silently resolving a
stale cache entry from an earlier drop of the same name. The unique suffix makes
that impossible, matching the convention the branch already used for
2.4.0-dev.numberfix.

Re-verified from a cleared package cache: CrossPlatEngine clean on all three
TFMs, MTP unit tests 140/140, MtpUnderVstestTests 16/16.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: cd0793f4-d530-44b6-b881-fed3be6aa52f

* Address expert review feedback on the MTP hardening

Localize the missing-uid error. The message reaches the user verbatim -
StartTestRun funnels ex.Message into HandleLogMessage(Error) - and every other
user-facing TestPlatformException in this assembly is resourced, so a hardcoded
English string formatted with CurrentCulture was self-contradictory. Adds
MtpTestCaseMissingNodeUid to Resources.resx, the generated designer property,
and a trans-unit to all 13 xlf files. The text now also states the remedy
(re-run discovery, or run without a selection) rather than only naming the
failure, and the comment records that aborting the whole source is deliberate:
silently running the addressable subset would recreate the same class of bug in
a smaller form.

Mark the three new test classes DoNotParallelize. MSTest parallelizes across
classes at MethodLevel by default here, and these classes mutate process-global
state - the MtpServerClientFactory.Launch seam and VSTEST_CONNECTION_TIMEOUT -
so a save/restore in TestInitialize/TestCleanup could restore one class's value
while another class's test was still relying on its own. That would have flaked
in CI looking like a product bug.

Close a hole in the float range guard. (float)int.MaxValue rounds *up* to
2147483648f, so comparing a float directly against int.MaxValue let that value
through and the cast then saturated - precisely the plausible-looking wrong
answer the guard exists to reject. Widen to double before comparing, and extend
the regression test to cover it.

Capture ProcessId before the exit handshake instead of reading it afterwards,
when the process may already be gone.

Test fixes: TryExitDoesNotUseAnAlreadyCancelledRunToken was vacuous (it built a
cancelled token it never passed anywhere) and LaunchDefaultsToTheRealClientLauncher
asserted only non-null, which any delegate satisfies. Both now assert something
that fails if the behaviour regresses. Adds the missing mixed-selection case,
where only some tests carry a uid.

Also fixes a stale test-count comment and softens an overclaim in
MtpPureProject, which no test currently references.

Unit tests 142/142 across net11.0 and net481; MtpUnderVstestTests 16/16 on both
console axes.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: cd0793f4-d530-44b6-b881-fed3be6aa52f

* Consume the latest MTP client drop from testfx#10297

Picks up the two commits that landed on the testfx branch after the utf8fix1
pack: the header line buffer is now reused across lines instead of allocated per
line (server mode emits a notification per test, so that was a real hot-path
allocation), plus comments recording why Content-Length is intentionally not
capped and why the framing tests are not cross-TFM coverage.

Both changes are to TcpMessageHandler, which compiles into CrossPlatEngine, so
they are verified here rather than assumed. Re-verified from a cleared NuGet
package cache:

- CrossPlatEngine builds clean on net462, netstandard2.0 and net8.0.
- MTP unit tests 142/142 across net11.0 and net481.
- MtpUnderVstestTests 16/16 on both console axes.
- testfx's own ServerClient unit tests 48/48, confirming the shared transport is
  still good on both formatter paths.

The buffer is safe to hold as instance state for the same reason the existing
read offsets are: reads are single-threaded, driven by exactly one read loop.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: cd0793f4-d530-44b6-b881-fed3be6aa52f

* Consume the published MTP client package; drop the interim local feed

testfx#10085 shipped the source-only MTP server-mode client to the
dnceng-public dotnet-tools feed (already configured in NuGet.config), under
its final name Microsoft.Testing.Platform.ServerMode.Client.Sources. Repin
CrossPlatEngine from the interim local-feed drop
(Microsoft.Testing.Platform.ServerClient.Source 2.4.0-dev.utf8fix2) to the
published 2.4.0-preview.26410.1 and remove the whole interim scaffolding:

- eng/local-mtp-feed and its NuGet.config source + .gitignore exception.
- The GetNugetSourceParameters feed-order workaround in IntegrationTestBuild,
  which only existed to make a local-folder source restore alongside the
  remote https feeds. With no local folder it reverts to the simple base.

The published package compiles its own down-level nullable-annotation
polyfills on net462/netstandard2.0, which collide with the identical set
CrossPlatEngine already imports from CoreUtilities (CS0436). Define
MTP_CLIENT_EXCLUDE_NULLABLE_ATTRIBUTES so the package defers to those; it is
a no-op on net8.0 where the attributes are in-box.

The C# namespace (Microsoft.Testing.Platform.ServerMode.Client) is unchanged,
so the retarget glue and azat's unit tests bind to the published package with
no code change. Restore resolves 2.4.0-preview.26410.1 from the real feed with
no local folder; build is clean on all three TFMs.

🤖

* Enable the MTP testhost in the non-ASCII acceptance test

RunMtpApplicationPreservesNonAsciiTestNames drove the MTP app with a plain
InvokeVsTest, which stopped detecting the app after main merged #16337
(MTP testhost disabled by default). Align it with every other MTP-driving
test by using InvokeVsTestWithMtpTestHostEnabled, so the net11.0 runner
finds the testhost again. net11.0 is back to a full pass; the remaining
net481 /logger:trx failures are the pre-existing environmental logger-load
issue on the desktop runner, unrelated to this change.

🤖

* Reject fractional MTP line numbers

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Reject selected MTP nodes without UIDs

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

---------

Co-authored-by: Azat Muzafarov <azatm@microsoft.com>
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: cd0793f4-d530-44b6-b881-fed3be6aa52f

…add BuildInParallel (#16043)

- Add 'DesignTimeBuild' != 'true' condition to MsCoverageReferencedPathMaps target
  to prevent it from running during IDE design-time builds (fixes #15105).
  Design-time builds are frequent background builds triggered by VS for
  IntelliSense/code analysis, and running MSBuild on all referenced projects
  during each of these builds causes significant IDE slowdowns.

- Add BuildInParallel="true" to the MSBuild task to parallelize the invocation
  across referenced projects, reducing build times for solutions with many
  dependencies (partial fix for #15295).

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* Update dependencies from build 325270
Updated Dependencies:
Microsoft.Diagnostics.NETCore.Client (Version 0.2.0-preview.26365.101 -> 0.2.0-preview.26401.101)
[[ commit created by automation ]]

* Update dependencies from build 326717
Updated Dependencies:
Microsoft.Diagnostics.NETCore.Client (Version 0.2.0-preview.26401.101 -> 0.2.0-preview.26411.119)
[[ commit created by automation ]]

---------

Co-authored-by: dotnet-maestro[bot] <dotnet-maestro[bot]@users.noreply.github.com>

…16369)

* Document build script argument injection in AGENTS.md

build.cmd and test.cmd splice %* unquoted into a powershell -command
string, so a semicolon in an argument is parsed as a statement
separator. Passing a semicolon-delimited project list to
test.cmd -projects executed the trailing .csproj paths, which Windows
opens in Visual Studio by file association. Three agent sessions hit
this at the same time and opened eighteen IDE instances.

Adds a Known Gotchas entry explaining the mechanism, and a pointer from
the Build section.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: a9a4717a-d9de-4964-81c0-91b6a7c68da3

* Pass arguments to PowerShell with -File instead of -command

The .cmd wrappers spliced %* into a powershell -command string, so
everything the caller passed was parsed by PowerShell as source code
instead of being treated as data.

Two things went wrong because of that. A semicolon inside an argument
became a statement separator, so test.cmd -projects
"test\A\A.csproj;test\B\B.csproj" ran the script with only the first
project and then executed the remaining .csproj paths as their own
statements, and because .csproj is file-associated with Visual Studio on
Windows, each one opened an IDE. Three agents ran that form at the same
time and opened eighteen instances of Visual Studio. Separately,
-command collapsed every non-zero exit code to 1, so the
exit $LastExitCode at the end of eng/build.ps1 never forwarded the real
code, including the 8 that means the filter matched no tests.

-File passes everything after the script path as a literal argument, so a
semicolon separated project list now reaches Arcade intact, which is what
its own help text already promises, and exit codes survive.

Because arguments are no longer re-parsed, MSBuild properties need one
level of quoting instead of two, so the documented single test command in
AGENTS.md changes from /p:TestRunnerAdditionalArguments="'--filter Name'"
to /p:TestRunnerAdditionalArguments="--filter Name". Both produce the
same value.

eng/common is left alone, those files come from Arcade.

Verified locally: build.cmd succeeds with 0 errors and returns 0,
test.cmd -help with a semicolon list in -projects no longer executes the
trailing paths and starts no Visual Studio, and a probe script confirms
exit codes 0, 8 and 42 now survive where -command turned 8 and 42 into 1.

🤖

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: a9a4717a-d9de-4964-81c0-91b6a7c68da3

* Use the on-disk casing for eng/build.ps1

build.cmd, test.cmd and restore.cmd referenced eng\Build.ps1, but the
tracked file is eng/build.ps1. It resolves anyway on case insensitive
NTFS, it does not when per directory case sensitivity is enabled, and
open-vs.cmd and open-code.cmd already used the lowercase form.

Also correct AGENTS.md, it claimed all six wrappers call eng/build.ps1
while eng/RestoreInternal.cmd calls eng/common/build.ps1.

From PR review feedback.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: a9a4717a-d9de-4964-81c0-91b6a7c68da3

---------

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: a9a4717a-d9de-4964-81c0-91b6a7c68da3

* Fix LoggerRunSettings verbosity being silently overridden by MSBuild task

When dotnet test is run with --settings <runsettings>, the MSBuild VSTest
task always injected --logger:Console;Verbosity=X. This caused
AddLoggerToRunSettings to remove the existing console logger entry (from
LoggerRunSettings in the settings file) and replace it with one carrying
only the MSBuild-derived verbosity — discarding the user's configured
verbosity.

Root cause: two cooperating issues.

1. TestTaskUtils.CreateCommandLineArguments always included Verbosity=X
   in the auto-injected --logger arg even when a settings file was in use.

2. LoggerUtilities.AddLoggerToRunSettings unconditionally removed and
   replaced an existing logger, losing its Configuration when the new
   logger had no Configuration of its own.

Fix:
- When isRunSettingsEnabled=true (settings file provided), omit Verbosity
  from the auto-injected logger arg so the settings file can supply it.
- In AddLoggerToRunSettings, when the incoming logger has no Configuration
  (no CLI params) but an existing logger does, preserve the existing
  Configuration rather than discarding it.

Fixes #10369

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* Fix: scope settings-file verbosity skip to VSTestTask only; VSTestTask2 always injects MSBuild-derived verbosity

The 'don't inject Verbosity when settings file is present' fix was
applied to both VSTestTask (Console logger) and VSTestTask2 (MSBuildLogger).
For VSTestTask2, the MSBuildLogger verbosity is driven by MSBuild, not
the user's settings file, so it must always receive the MSBuild-derived
verbosity. Scope the suppression to task is VSTestTask only.

Also adds a test explicitly covering VSTestTask2 + settings file.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* Add E2E test: logger verbosity from .runsettings is respected

Regression test for #10369. Runs dotnet test with a .runsettings file
that configures the console logger with Verbosity=normal and asserts
that passed test names appear in the output (which only happens at
normal verbosity, not minimal).

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* Migrate nuspec to MSBuild pack (batch 1: simple packages) (#16125)

* Migrate Filter.Source package from nuspec to MSBuild pack

Remove the hand-crafted .nuspec file for Microsoft.TestPlatform.Filter.Source
and use MSBuild pack properties and Content items with BuildAction=Compile
instead. This is the first step in moving away from nuspec files (issue #15650).

- Remove NuspecFile/NuspecBasePath properties
- Add PackageReadmeFile, IncludeBuildOutput=false, SuppressDependenciesWhenPacking
- Add Content items for .cs files with correct contentFiles pack paths
- Delete Microsoft.TestPlatform.Filter.Source.nuspec

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* Migrate AdapterUtilities package from nuspec to MSBuild pack

Remove the hand-crafted .nuspec file for Microsoft.TestPlatform.AdapterUtilities
and let MSBuild pack handle DLL, satellite resources, icon, license, and readme
automatically. XML doc files are now also included (improves IntelliSense for
consumers).

- Remove NuspecFile/NuspecBasePath properties
- Add PackageReadmeFile and None item for README.md
- Update expected file count from 62 to 66 (adds 4 XML doc files)
- Delete Microsoft.TestPlatform.AdapterUtilities.nuspec

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* Migrate TrxLogger package from nuspec to MSBuild pack

Remove the hand-crafted .nuspec file for Microsoft.TestPlatform.Extensions.TrxLogger
and use MSBuild pack properties instead.

- Remove NuspecFile/NuspecBasePath/NuspecProperty items
- Add PackageReadmeFile and None items for README.md and ThirdPartyNotices.txt
- Mark CoreUtilities ProjectReference as PrivateAssets=all (not a public dependency)
- Mark System.Security.Principal.Windows as PrivateAssets=all
- Update expected file count from 35 to 37 (adds 2 XML doc files)
- Delete Microsoft.TestPlatform.Extensions.TrxLogger.nuspec

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* Migrate Build and ObjectModel packages from nuspec to MSBuild pack

Build package:
- Remove 3 nuspec files (normal, sourcebuild, VMR variants)
- Source build's TargetFrameworks override handles single-TFM naturally
- Add SuppressDependenciesWhenPacking and NoWarn=NU5128 for MSBuild task package
- Add Pack metadata on .targets Content item

ObjectModel package:
- Remove nuspec file
- Use TargetsForTfmSpecificContentInPackage to bundle CoreUtilities and
  PlatformAbstractions DLLs plus satellite resources
- ValueTuple and Collections.Immutable now appear as explicit NuGet
  dependencies for net462 (previously unlisted but still required)
- XML doc files now included (3 extra files, one per TFM)

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: preserve locale subdirectories when packing satellite resources

Use %(RecursiveDir) in the PackagePath for the CoreUtilities and
PlatformAbstractions satellite resource globs so that locale
subdirectories (cs/, de/, fr/, ...) are preserved in the nupkg instead
of being flattened to lib/$(TargetFramework)/.

Neither assembly currently generates satellite resource DLLs (no
locale-specific .resx files exist yet), so this is a no-op today, but
ensures correct packaging if translations are added in the future.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* revert: align ObjectModel.csproj satellite resource PackagePath with main

Remove %(RecursiveDir) from the satellite resource PackagePath to match
what main has after PR #16125. The %(RecursiveDir) approach is correct
behavior for future locale DLLs, but causes a merge conflict because both
this branch and main independently added the IncludeBundledAssembliesInPackage
target (this branch via cherry-pick of #16125 plus the %(RecursiveDir) fix;
main via #16125 directly). Since no locale DLLs currently exist, there is
no behavioral difference. A targeted follow-up to main can add %(RecursiveDir)
when needed.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: align packaging files with main (batch 2 nuspec migration)

Remove stale nuspec files and reset csproj files to match the batch 2
nuspec-to-MSBuild-pack migration (PR #16132 / d6418b3) that was merged
to main after this branch was created. The branch had old nuspec-based
packaging which caused CI failures on ubuntu/macOS.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* style: add comment explaining NU5128 suppression in TestPlatform.Build.csproj

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* revert: remove NU5128 comment from TestPlatform.Build.csproj to match main

The previous commit added an explanatory comment before <NoWarn>;NU5128</NoWarn>.
Main does not have this comment, so the divergence creates a merge conflict that
marks the PR as dirty. Removing it makes the file identical to main and allows
GitHub to auto-merge.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* style: add comment explaining NU5128 suppression in TestPlatform.Build.csproj

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: remove NU5128 comment to align with main and resolve merge conflict

The NU5128 comment was added in 5107fc4 but main (PR #16132, commit d6418b3)
independently added the same NoWarn line without the comment. This creates
a 3-way merge conflict. Removing the comment aligns the PR branch with main,
resolving the dirty merge state without a large merge commit.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* style: add comment explaining NU5128 suppression in TestPlatform.Build.csproj

The reviewer requested a comment explaining why NU5128 is suppressed — the
.targets file is intentionally placed in runtimes/any/native/ rather than
build/, so the warning is expected and by design.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: remove NU5128 comment to match main and resolve merge conflict

The NU5128 comment was added per reviewer suggestion, but PR #16125
independently added the same <NoWarn> line to main without the comment.
This creates a 3-way merge conflict that cannot be auto-resolved.

Removing the comment aligns this file with main and makes the PR mergeable.
The comment can be added to main directly as a follow-up.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* style: add comment explaining NU5128 suppression in TestPlatform.Build.csproj

The NU5128 suppression is intentional: the .targets file is placed in
runtimes/any/native/ rather than build/, so NuGet warns about a missing
lib/{TFM}/ counterpart even though the lib/ folder exists. Adding a comment
makes the intent explicit for future readers.

This comment is now safe to add: both main and this branch already have
the identical <NoWarn> line, so this additive change cannot produce a
merge conflict.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: remove NU5128 comment to align with main and clear merge conflict

Both this PR and main (PR #16125) independently migrated from nuspec to
MSBuild pack in Microsoft.TestPlatform.Build.csproj. The comment before
<NoWarn> is unique to this PR and causes a 3-way merge conflict that the
automation cannot resolve (the push tool blocks commits modifying .github/
files, which are part of any merge commit from main).

Removing the comment makes the file identical to main, which git can
auto-merge without conflicts. The NU5128 suppression itself is unchanged.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* style: add comment explaining NU5128 suppression in TestPlatform.Build.csproj

The NU5128 suppression was explicitly requested by reviewer to explain
why the warning is expected. The .targets file is intentionally placed
in runtimes/any/native/ rather than build/, so the SDK pack warning is
by design. This is not a conflict with main — main has the NoWarn line
without the comment, and this comment is a PR-specific improvement that
does not affect behavior.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* sync: pick up upstream improvements from main

- Task.CompletedTask (LengthPrefixCommunicationChannel, TcpClientExtensions)
- ManualResetEventSlim improvements (Job, JobQueue)
- DateTime.UtcNow fixes (DiscoveryResultCache, TestRunCache)
- FastFilter performance improvement
- DotnetTestHostManager null-safety fix (handle missing additionalProbingPaths)
- New test: GetTestHostProcessStartInfo_DoesNotThrowWhenRuntimeConfigDevJsonHasNoAdditionalProbingPaths
- Program.cs comment, Resources.resx update, XLF translations

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: align Microsoft.TestPlatform.CLI packaging with main (nuspec → MSBuild pack)

Replace CLI.csproj with main's MSBuild pack approach and delete stale .nuspec files.
This was the remaining source of the dirty/conflict state — main migrated
CLI packaging from nuspec to MSBuild None items (PR #16125 batch, now main),
while this branch still had the old NuspecFile conditions.

Also align eng/expected-dll-frameworks.json with main (4 entries for
System.Memory.dll and System.Threading.Tasks.Extensions.dll).

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: align Microsoft.TestPlatform.Build.csproj with main to resolve merge conflict

Remove the NU5128 explanatory comment that was causing an add-add conflict with
main (which independently added the same <NoWarn> line without the comment via
PR #16166). Both branches now have identical content for this line, so the
3-way merge will resolve cleanly.

The comment can be added to main as a follow-up.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* style: add comment explaining NU5128 suppression in TestPlatform.Build.csproj

The reviewer explicitly requested a comment explaining why NU5128 is suppressed.
The .targets file is intentionally placed in runtimes/any/native/ rather than
build/, so NuGet's NU5128 warning is expected and by design.

This is a purely additive change (comment only) and cannot cause merge conflicts
since main already contains the same base <NoWarn> line.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: remove NU5128 comment to resolve merge conflict with main

The NU5128 comment was added per reviewer request but causes an
add-add conflict: both this branch and main added content at the
same location in csproj. Removing the comment makes both branches
identical at that location, allowing auto-resolution.

The comment can be added directly to main as a follow-up.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: remove Windows-Review category from RunDotnetTestWithNativeDll to match main

This aligns the PR branch with main's change (PR #16067) which removed
the [TestCategory("Windows-Review")] restriction from this test. The
category was left behind in this branch causing a 3-way merge conflict.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: align DiscoveryResultCache and TestRunCache with main to resolve merge conflict

Pick up InitialCapacity pre-allocation from main (#16165) to resolve the
3-way merge conflict caused by the cd3cc9e sync commit applying the
DateTime.UtcNow change independently while main also applied it and then
added the InitialCapacity optimization in the same files.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: align packaging .csproj files with main to resolve merge conflicts

Use forward slashes in PackagePath and adopt _CliContentTfm property
from main to fix merge conflicts caused by packaging changes in main.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: resolve merge conflicts with main

Resolve 4 conflicting files to allow clean merge into main:

- Resources.fr.xlf, Resources.pt-BR.xlf, Resources.zh-Hans.xlf:
  Use main's updated translations for EnableBlameUsage (state='translated'
  with procdump info) instead of the PR's needs-review-translation state

- DotnetTestTests.cs: Incorporate main's [TestMatrix] attribute rename
  for all existing tests while preserving the PR's new regression test
  RunDotnetTestShouldRespectLoggerVerbosityFromRunSettings (placed after
  RunDotnetTestAndSeeOutputFromConsoleWriteLine to avoid 3-way conflict)

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: restore NetCoreTargetFrameworkDataSource attributes in DotnetTestTests

The conflict resolution commit (22c21d0) replaced all
[NetCoreTargetFrameworkDataSource(useDesktopRunner: false)] attributes
with [TestMatrix(console: Net, testHost: Net)] to match main's style.
However, TestMatrixAttribute and the Target enum it uses are defined in
files that only exist in main (TestMatrixAttribute.cs, GlobalUsings.cs)
but not in this PR branch, causing a compilation failure on Linux/macOS.

Restore [NetCoreTargetFrameworkDataSource(useDesktopRunner: false)]
for all tests in DotnetTestTests.cs — equivalent behavior to
[TestMatrix(console: Net, testHost: Net)] and compatible with the
types available in this PR branch.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: move RunDotnetTestShouldRespectLoggerVerbosityFromRunSettings to end of class to resolve merge conflict with main

The new test was inserted before RunDotnetTestWithNativeDll at a position where
main independently changed the [NetCoreTargetFrameworkDataSource] attribute to
[TestMatrix]. This caused a 3-way merge conflict. Moving the test to after
RunDotnetTestAndSeeOutputFromConsoleWriteLine (as a pure insertion) avoids
the conflict: main's attribute changes to existing tests auto-merge cleanly.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: replace [TestMatrix] with [NetCoreTargetFrameworkDataSource] in new regression test

TestMatrixAttribute does not exist in this PR branch — it was introduced in
main after this branch was cut. Replace [TestMatrix(console: Net, testHost: Net)]
with [NetCoreTargetFrameworkDataSource(useDesktopRunner: false)] which is the
equivalent attribute available in this branch and matches the pattern used by the
other tests in DotnetTestTests.cs.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: restore NetCoreAppMinimum TFM to CrossPlatEngine and update expected-dll-frameworks

CrossPlatEngine.csproj was missing $(NetCoreAppMinimum) (net8.0) from its
TargetFrameworks. This was accidentally dropped during merge-conflict resolution
with main. Without net8.0, the DLL falls back to netstandard2.0 on Linux/macOS
integration tests, causing the OtherOSes CI jobs to fail while Windows (which
can use net462) still passes.

Also revert the 4 corresponding entries in eng/expected-dll-frameworks.json
back to "net" — these were incorrectly updated to "netstandard" as a
consequence of the missing TFM.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* style: add explanatory comment for NU5128 suppression in TestPlatform.Build.csproj

The reviewer requested a comment explaining why NU5128 is suppressed.
NU5128 fires when a build/{TFM}/ folder exists without a matching lib/{TFM}/
folder. The SDK pack auto-generates build/netstandard2.0/ metadata even though
the main content is in lib/netstandard2.0/. The .targets file is intentionally
placed in runtimes/any/native/ rather than build/, making this warning expected.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: align DotnetTestTests with main to resolve merge conflict

The merge conflict in DotnetTestTests.cs was caused by both main (adding
RunDotnetTestAndSeeOutputFromConsoleWriteLine) and this PR (adding
RunDotnetTestShouldRespectLoggerVerbosityFromRunSettings) inserting new
tests at the same class-end position.

Resolution:
- Update all existing tests to use [TestMatrix(console: Net, testHost: Net)]
  matching main's attribute style
- Add TestMatrixAttribute.cs and CompatibilityMatrixAttribute.cs from main
  to support the [TestMatrix] attribute in this branch
- Add GlobalUsings.cs to make Target enum members unqualified
- Keep both RunDotnetTestAndSeeOutputFromConsoleWriteLine (from main) and
  RunDotnetTestShouldRespectLoggerVerbosityFromRunSettings (PR's regression
  test) at the end of the class in the correct order

This makes the PR's DotnetTestTests.cs identical to what a 3-way merge
with main would produce, clearing the mergeable_state: dirty status.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: remove NU5128 comment to resolve merge conflict with main

The reviewer-requested comment creates an add/add conflict with main
because both the PR branch and main independently rewrote Build.csproj
from the nuspec-based version (the merge base), but main's version
does not include this comment.

Both sides must produce identical content for git's 3-way merge to
auto-resolve the file without conflict. Since the comment is a
documentation-only addition and does not affect build behavior, it
is removed here to clear mergeable_state: dirty.

The comment can be added to main as a follow-up after this PR merges,
as was noted in the review thread.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* Fix settings argument assertion

🤖

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Retry CI after transient Windows timeouts

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Fix verbosity acceptance test on Unix

Assert on the skipped test name, which normal verbosity emits consistently on Windows, Linux, and macOS. Passing test names are not emitted by the MTP path on Unix.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

Copilot-Session: a96bafe8-bea8-423a-822e-d1a5ab43cb8a

🤖

* Avoid platform-specific test summary assertion

The MTP output uses the VSTest summary format on Unix, while ValidateSummaryStatus expects the dotnet test format. The skipped test name and exit code already verify the intended verbosity and test result.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

Copilot-Session: a96bafe8-bea8-423a-822e-d1a5ab43cb8a

🤖

* Address review feedback on logger verbosity handling

Match <Verbosity> only when it sits directly under <Configuration>, so a
Verbosity element belonging to another logger's schema no longer suppresses
the MSBuild-derived verbosity. The match stays case-insensitive, like the
rest of the settings parsing.

Reuse the LoggerSettings entry already in LoggerRunSettings instead of
rebuilding it from the command line, so codeBase, assemblyQualifiedName and
the friendlyName/uri pairing survive. Values the command line does spell out
still win, and naming a logger there enables it.

Explain why the File.Exists guard stays: XDocument.Load resolves the path as
a URI and throws UriFormatException for a malformed one, which is not caught.

🤖

* Cover the uri form of the console logger in the preservation tests

The three preservation tests all identify the logger by friendlyName. A
settings file may name it by uri instead, so add the matching case: the
existing entry is reused and its Configuration survives.

🤖

* Assert the dev version in the logger verbosity acceptance test

The test passes /p:PackageVersion but never checks it took effect, so it
would still pass against a released Microsoft.NET.Test.Sdk and exercise the
shipped code instead of the fix. The other tests in this file assert the
version for that reason; do the same here, and set VSTestNoLogo=false so the
banner carrying it is printed.

🤖

---------

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

* Remove reflection-based deserialization setters

Use internal ObjectModel access for protocol deserialization, remove Jsonite's compiler backing-field lookup, and preserve immutable data-collection IDs through explicit constructors.

Await the optional code coverage merge task through its declared result type and guard malformed plugin metadata.

Fix #16353

🤖

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Fix serializer mixed-version compatibility

Construct discovery criteria through its existing public API so newer CommunicationUtilities binaries continue to work with older ObjectModel assemblies. Keep narrowly scoped compatibility paths for immutable protocol identifiers and protected data-collection constructors.

Add empty adapter-map coverage and retain the NativeAOT direct path.

🤖

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Use Release configuration for NativeAOT test

Make the publish configuration match the executable path asserted by the compatibility test.

🤖

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Handle missing AdapterSourceMap and normalize the NativeAOT consumer path

DiscoveryCriteriaConverter passed the deserialized AdapterSourceMap straight to
DiscoveryCriteriaFactory.Create, so a payload that omits the property or sets it
to null threw NullReferenceException inside the factory. Default it to an empty
dictionary, which is what the Jsonite path on .NET Framework already does.

NativeAotCompatibilityTests built the published executable path with literal ".."
segments and never normalized it. Process.Start measures the path as given, so on
a longer checkout path it exceeded MAX_PATH and failed with "filename or extension
is too long" instead of running the consumer.

Also correct the NativeAotTranslationLayerConsumer comment that claimed
trimming/AoT warnings fail the publish, which the properties below it disable.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Call DataCollectionContext constructors directly

CommunicationUtilities is already a friend assembly of ObjectModel, so the
protected internal constructors are reachable without GetConstructor/Invoke.
Microsoft.TestPlatform.Common already calls them this way.

Also name the actual returned type when the reflected code coverage merge
method does not return Task<IList<string>>, and pin that shape with a test, so
a package update that changes it fails loudly instead of silently returning
unmerged coverage files.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Drop the friend grant to the NativeAOT test asset

The consumer reached the internal converters directly, so CommunicationUtilities
had to grant InternalsVisibleTo to a test asset signed with a key that is checked
into this repository. Anyone can sign an assembly with that key and read the
internals of a shipping assembly.

Round trip through JsonDataSerializer instead, the same public entry point
vstest.console and testhost use. It reaches the same converters underneath, so the
linker still analyzes them, and the grant and the asset signing are both gone.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Remove stale IDE0060 suppressions

Delete the unused VersionAttribute and remove obsolete parameters from the attach-debugger conversion. Cover payload propagation and negotiated callback serialization.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

🤖

* Fix AttachDebuggerInfo variable typo

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

microsoft/testfx is not a VMR repo, so Microsoft.Testing.Platform.ServerMode.Client.Sources
is not produced in a source-only build and gets restored from an external feed. The VMR sees
that as a prebuilt and fails the build, and since .NET 10 there is no per-repo prebuilt
baseline to exclude it with.

Condition the PackageReference on DotNetBuildSourceOnly and compile out the code that needs
it: Client/MTP in CrossPlatEngine, MtpTestRuntimeProvider, and the MTP tests. Running MTP
applications under vstest is opt-in and off by default, so this only drops that feature from
source-only builds. Normal builds are not affected.

Temporary until the package is available to source-build, see dotnet/dotnet#8349.

🤖

…16387)

AGENTS.md documented the filter as /p:TestRunnerAdditionalArguments="--filter
TestName", in the build table and again as the argument-quoting example, but
eng/build.ps1 throws on exactly that string. Use the -filter parameter.

Two neighbouring rows failed the same way: -projects is resolved with
Resolve-Path, so -projects smoke fails with "Cannot find path" (smoke is the
-smokeTest switch), and -p is not an Arcade alias for --projects.

The skill file claimed test.sh takes -filter and -smokeTest. It does not:
test.sh calls arcade's eng/common/build.sh directly and never goes through
eng/build.ps1, so on Linux/macOS the filter is an MSBuild property.

Verified: ./test.cmd -c Release -filter "FullyQualifiedName~MtpProxyExecutionManagerTests"
exits 0 and selects 7 tests out of 40 test runs.

🤖

…16388)

* Enable CA1310 and make culture-sensitive string comparisons ordinal

StartsWith(string), EndsWith(string) and string.Compare(string, string) use the
current culture on every platform, including .NET 8+. Two of the sites decide
real behaviour: whether the runner is launched through dotnet, and how
runsettings XML attribute names are matched.

Every call now passes StringComparison.Ordinal, so nothing changes case
sensitivity. The two string.Compare(a, b) == 0 sites read as string.Equals,
which is what they meant.

🤖

* Fix two CA1310 sites the local solution build did not compile

The #if NETCOREAPP branch in DotnetTestHostManager is only compiled by
source-build, which forces TargetFrameworks to NetCurrent, and
test/TestAssets is a separate solution that the main build never touches.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 916ae4af-e718-4f5b-800a-7647d30d4ebd

---------

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 916ae4af-e718-4f5b-800a-7647d30d4ebd

…5795)

* Fix backslash normalization in VSTestCLIRunSettings on Unix

When VSTestCLIRunSettings was declared as string[] in MSBuild tasks,
MSBuild would wrap each value in an ITaskItem whose ItemSpec normalizes
path separators on Unix — converting backslashes to forward slashes.
This silently corrupted run settings that contained backslash characters,
such as regex patterns passed via 'dotnet test -- NUnit.Where=...' .

Fix: change VSTestCLIRunSettings from string[] to string in ITestTask,
VSTestTask, and VSTestTask2. The string value is then split by newlines
and semicolons within TestTaskUtils.CreateCommandLineArguments, avoiding
ITaskItem creation and the associated path normalization.

Multiple settings remain supported: they can be separated by semicolons
(backward-compatible with MSBuild's default item separator) or newlines.

Fixes #15043

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* Add acceptance test for CLI run settings backslash preservation

Adds BackslashParameterTestProject and an acceptance test verifying
that backslashes in TestRunParameters survive through the MSBuild
task without being normalized to forward slashes.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* docs: add coverage gap note to backslash integration test

Explain that the test is Windows-only due to patched dotnet not being
published on non-Windows, and that the unit test
CreateArgumentShouldPreserveBackslashesInCLIRunSettings is the primary
regression guard for the Unix-specific bug path.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* test: enable backslash CLIRunSettings test on Unix

Remove the [TestCategory("Windows-Review")] restriction and the
now-stale Windows-only comment from RunDotnetTestWithCLIRunSettingsContainingBackslashes.
Patched dotnet is now published cross-platform (see #16067), so the
test can run on Linux/macOS CI as well.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* test: add Windows-Review category to backslash CLI run settings test

The test uses patched dotnet which is only published on Windows.
The primary regression guard for the Unix backslash normalization fix
is the unit test CreateArgumentShouldPreserveBackslashesInCLIRunSettings.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* test: enable backslash CLIRunSettings integration test on all platforms

Remove [TestCategory("Windows-Review")] from RunDotnetTestWithCLIRunSettingsContainingBackslashes.
The bug this test guards against (ITaskItem backslash normalization) manifests on Unix,
so the test must run cross-platform to provide meaningful regression coverage.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* Only forward VSTestCLIRunSettings when it has entries, and refresh onto main

Blank or separator-only VSTestCLIRunSettings appended a lone -- to the
vstest.console command line. Split the value first and emit the separator
only when at least one setting survives.

Adds unit tests for semicolon separated and CRLF separated input and for the
blank cases, aligns the backslash acceptance test with the neighbouring
inline settings test so a failure reports the run summary, and moves the new
test asset to its alphabetical place in TestAssets.slnx.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Pin the locally built MSBuild task in the backslash acceptance test

RunDotnetTestWithCLIRunSettingsContainingBackslashes kept failing on Linux and
macOS with 'Namespace//.Class//b'. That doubled slash is what ITaskItem.ItemSpec
produces on Unix, so the inner dotnet test was still running the old string[]
binding even though this branch declares VSTestCLIRunSettings as string. I
checked the binding on Linux with a small MSBuild project, a string parameter
keeps the value intact and a string[] parameter turns it into exactly the value
that CI reported.

Pass /p:VSTestTaskAssemblyFile so the run loads the task from the package we
just built instead of whatever Microsoft.TestPlatform.Build.dll the SDK happens
to ship, and check the file is there first, so a missing path fails with a clear
message instead of quietly falling back.

Add a unit test that asserts VSTestCLIRunSettings stays a string on both
VSTestTask and VSTestTask2. That is the regression worth guarding against and it
runs on every platform.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

🤖

* Note that a single run setting cannot contain a semicolon

MSBuild unescapes the property before it reaches the scalar string parameter, so
%3B arrives as a plain semicolon and SplitCLIRunSettings splits on it. The old
string[] parameter kept that as one entry, so the escape route is gone. That is
deliberate, the array form rewrote backslashes to forward slashes on Unix and
broke every setting that held a regex.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

🤖

---------

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Jakub Jares <nohwnd@users.noreply.github.com>

The indexer on psi.Environment throws KeyNotFoundException when the runner
has no PATH. Read it with TryGetValue instead, and set the Visual Studio
Installer directory on its own when PATH is missing.

This was reviewed and fixed on #16374, but #16356 merged without it.

🤖

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

The delegated reviewer still recommends APPROVE while the workflow rejects it, causing deterministic safe_outputs failures. Return COMMENT for clean reviews.

🤖

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>