Skip to content

0.48.2-RC1 - #3860

Open
icanhasmath wants to merge 90 commits into
betafrom
marcg/0-48-2-RC1
Open

icanhasmath wants to merge 90 commits into
betafrom
marcg/0-48-2-RC1

Conversation

@icanhasmath

Copy link
Copy Markdown
Contributor

Beta release 0.48.2-RC1 (PROD-240).

This is master as of 44e750f plus the changelog. master follows on from the 0.48.1 beta merge-back (904939d, Merge branch 'version/0-48-2-RC1'), and the changes merged since then are:

version.txt already reads 0.48.2-RC1. The commit count against beta is large because the RC branches reach master as merges with different hashes. Only the three PRs above are new.

Merging this stages the release. After that come Windows signing, state run promote-staging-to-production, verifying with install.sh -b beta, and merging back into master.

Changelog

Added

  • state publish --build now reads [project].dependencies from pyproject.toml and publishes them as runtime
    dependencies of the encrypted private ingredient. Version ranges carry over, and OS and Python version markers become
    dependency conditions (unsupported markers are skipped with a notice).
  • Added state export ingredient-meta <namespace>/<name>[@version], which writes an existing ingredient's meta file for
    use with state publish --edit --meta.

Changed

  • Installing runtimes on Windows is significantly faster: link paths are resolved once per tree instead of once per
    file, and artifacts are linked concurrently.
  • A --meta file's name, version, and description now feed the built wheel (flags take precedence over the meta file,
    which takes precedence over pyproject.toml).

Fixed

  • state publish --edit no longer overwrites the meta file's dependencies and authors with the previous revision's,
    keeps inherited dependency types and conditions, and rejects author or description changes on an existing version
    up front.
  • state publish --edit --meta with an exported meta file now edits the version named in the file rather than the
    latest version.

Security

Naatan and others added 30 commits September 15, 2025 14:21
Recognize hyphenated directory names in some Rust crates.
Recognize planning errors when evaluating a build expression.
After dynamically evaluating a build plan, staging a commit needs to use the latest timestamp.
Co-authored-by: Nathan Rijksen <nathanr@activestate.com>
Fixed failing CVE integration test.
Ecosystem namespaces will contain a mix of dynamically-imported artifacts and built artifacts.

DI artifacts produced by the star-builder should invoke an ecosystem-specific installation procedure.

Artifacts produced by other builders should be installed as normal.

Note: uninstallation will not work because depot artifacts are not tied to builders. This will need to be addressed in a follow-up ticket.
Replace mholt/archiver with mholt/archives.
mitchell-as and others added 29 commits January 7, 2026 11:09
Skip uninstall integration test on Windows.
… WebSocket

The build-log-streamer WebSocket previously opened with only an Origin header, so the server had no way to authorize the stream. When the caller is authenticated, offer the platform JWT via Sec-WebSocket-Protocol as 'bearer.<jwt>' alongside the real 'build-log-streamer.activestate.com.v1' subprotocol the server echoes back (its allow-list contains only the real subprotocol, so the token never appears in the upgrade response). The browser WebSocket API can't set custom request headers, so the dashboard carries the JWT the same way -- both clients stay symmetric.

Also send the versioned State Tool User-Agent on the Upgrade so the server can see which client versions are connecting.

The token is threaded as a plain string from the runtime options into Connect, so pkg/runtime gains no authentication dependency; an empty token means anonymous (unchanged behavior). Includes a real-handshake test that asserts the offered subprotocols, the negotiated subprotocol, and the User-Agent the server receives.
State Tool: forward platform JWT + version User-Agent on the build-log-streamer WebSocket
* Make serve-org-keys work in cmd shell as well as bash

Add a batch twin of serve-org-keys constrained to cmd, and constrain
the existing bash version to non-cmd shells, so the script resolves per
shell the same way install-deps-dev does.

* Run in cmd without openssl or bash

Add --gen-cert to orgkeyserver.py so it generates its own self-signed
cert/key via the cryptography package, removing the openssl dependency
that on Windows only comes from git-bash. Update the cmd/batch variant
to detect the Python interpreter (python, py -3, python3) and pass
--gen-cert instead of shelling out to openssl.

- Quote --org in the no-key branches (bash and batch) so org names with
  spaces or shell metacharacters are passed as data, matching the
  --key branches.
- chmod the generated TLS private key to 0600 so it isn't world-readable.

* Drop openssl entirely and have orgkeyserver.py always generate its self-signed cert.

- Add CERT_DIR="test/ssl" constant; generate_self_signed(host) writes
  cert.pem/key.pem there and returns their paths.
- Remove --tls-cert/--tls-key/--gen-cert; cert generation is now the
  default on every run.
- Assume cryptography is present (bundled in the runtime); drop the
  ImportError guard.
- Simplify both activestate.yaml variants to just run the script with
  --org (and optional --key); remove the openssl/mkdir prep.

Resolve the cert dir to an absolute path in
generate_self_signed(), and print the absolute key_service_ca path on
startup so the value is copy-pasteable regardless of cwd. Docstring
example updated to an absolute /path/to/test/ssl/cert.pem placeholder.

* Bump project runtime commitID

Update to a runtime that bundles cryptography, which serve-org-keys'
orgkeyserver.py relies on for cert generation.

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Add environment-variable overrides for all config options

Every registered config option can now be overridden by an environment
variable, so config can be applied machine-wide (all users) without a
shared config file and without sharing auth.

Each option gets a canonical override derived from its key, e.g.
"update.info.endpoint" -> ACTIVESTATE_CONFIG_UPDATE_INFO_ENDPOINT. The
five options that already had bespoke env vars (ACTIVESTATE_API_HOST,
ACTIVESTATE_CLI_ANALYTICS_PIXEL, ACTIVESTATE_NOTIFICATIONS_OVERRIDE,
ACTIVESTATE_TEST_UPDATE_URL, ACTIVESTATE_TEST_UPDATE_INFO_URL) keep
working as aliases; the canonical variable takes precedence when both
are set.

Overrides are resolved in config.Instance.Get (env > stored > default)
so they actually take effect everywhere, not just in display. Values are
coerced to the option's type. Only registered options are eligible, which
keeps credentials (e.g. apiToken, not a registered option) out of scope.

`state config` now reports the effective value and a Source column
showing the overriding env var name, "local", or "default"; JSON output
gains source, env, and envVar fields.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* Address PR review: validate env override values; add locale strings

- EnvOverride now strictly coerces env values to the option's type and
  ignores invalid ones (unparseable bool/int, out-of-set enum), so a
  mis-typed variable falls back to the stored/default value instead of
  silently becoming a zero value. Mirrors the validation `state config
  set` already performs.
- Add the config_source_local and config_source_default locale strings
  used by the Source column instead of relying on the fallback path.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* Fix TestList integration test for new Source column

The `*` marker that denoted a locally-set value was replaced by the
Source column, so `state config` now shows "optin.buildscripts true
local" instead of "true*". Update TestList to assert the "local" source
(and the new "Source" header) instead of the removed "true*" marker.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* Add integration test for environment as config source

TestListEnvSource overrides optin.buildscripts via its canonical env var
(ACTIVESTATE_CONFIG_OPTIN_BUILDSCRIPTS) without running `state config
set`, then asserts the override takes effect and is reported as coming
from the environment.

The human-readable table confirms the non-default value; the source is
asserted via JSON output ("source":"environment" plus the overriding env
var name), which is robust against the table wrapping long variable
names.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* Address review feedback from mitchell-as

- Fold RegisterOptionWithEnv into RegisterOption via a variadic
  envAliases parameter; drop the separate function. Keep the note that
  envAliases are only for legacy env vars without the ACTIVESTATE_CONFIG_*
  prefix.
- Use real, user-facing config options in doc/test examples (api.host,
  security.prompt.level) instead of the test-only update endpoints.
- Trim the Instance.Get comment to just note env precedence; drop the
  machine-wide framing.
- Remove the unnecessary comment in config get.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
state-remote-installer.exe was triggering a Windows UAC elevation prompt;
when an admin authenticated it, the State Tool installed under the admin
account instead of the invoking user. The binary shipped with no embedded
application manifest, so it declared no requestedExecutionLevel and was
subject to Windows' UAC installer-detection heuristic (which auto-flags
executables whose filename contains "install" for elevation).

Add a single shared application manifest declaring asInvoker/uiAccess=false
and embed it into every State Tool executable via goversioninfo:

  - state, state-svc: ManifestPath added to the shared versioninfo.json
    template (internal/assets/contents), which is stamped into each
    binary's versioninfo.json at preprocess time.
  - state-remote-installer: committed versioninfo.json repointed to the
    shared manifest (replacing the ad-hoc local manifest).
  - state-installer, state-exec, state-mcp: had no resource embedding at
    all; added //go:generate goversioninfo, versioninfo.json generation in
    preprocess, and a Windows-guarded `go generate` in their build steps.

This opts every binary out of installer-detection auto-elevation so they
run with the invoking user's token. Verified by cross-building each cmd
for windows and confirming the asInvoker manifest is present in the
embedded resource.syso.

Note: this is prevention for new installs. A "Run as administrator" choice
already cached under AppCompatFlags\Layers (RUNASADMIN) overrides the
manifest and must be cleared separately on affected machines.

CS-2357

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
…ler-uac

CS-2357-Declare asInvoker execution level for remote installer
* Resolve link paths once per tree instead of once per file

smartlink.Link resolved both src and dest via ResolveUniquePath for every
entry it recursed into. On Windows ResolveUniquePath calls GetLongPathName,
which is a syscall, so installing a runtime with many files performed two
extra syscalls per file - a major contributor to slow Windows installs.

The tree root is already resolved by LinkContents (and now by the public
Link), and descendant paths are built by joining an already-resolved parent
with real (long) entry names, so they need no further resolution. Split the
recursion into an unexported link() worker that assumes resolved inputs and
does not re-resolve, keeping resolution only for the rare symlink-to-file
case where the pre-existing behavior links to the dereferenced target.

Also treat an "already exists" error from linkFile the same as the existing
TargetExists skip, so concurrent deploys of a shared file degrade to the
documented "first one wins" behavior rather than failing.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* Deploy linked artifacts concurrently instead of serializing them

DeployViaLink held the depot-wide fsMutex for the entire link operation, so
even though the install worker pool runs several artifacts at once, the
actual file linking was serialized to one artifact at a time. On Windows,
where each link is an individual (relatively slow) syscall, this made the
install phase far slower than it needed to be.

Linking touches no shared in-memory depot state (Exists guards its own map
via mapMutex and depotPath is immutable after construction), directory
creation is idempotent, and smartlink now treats an already-existing target
as a skip, so it is safe to run without the lock. Drop it from DeployViaLink
so linking parallelizes across the worker pool.

DeployViaCopy keeps the lock: two artifacts copying the same file would race
on os.Create and could corrupt the destination. That path is only used for
artifacts needing file transforms, when hard links are unsupported, or in
portable mode, so it remains the less common case.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* Hold a read lock while linking so copies can't write through links

Linking without the depot lock let a concurrent copy deployment write
through a link created moments earlier: copyFiles checks the destination
is free, a link to another artifact's depot file appears there, and
os.Create then truncates and overwrites that depot file, corrupting the
shared cache. TestConcurrentCopyAndLinkKeepDepotIntact reproduces it in
every run without the lock.

Make fsMutex an RWMutex: DeployViaLink holds it for reading, so links
still run concurrently, while copies, Put and Undeploy hold it for
writing. Also fix the non-constant format string in DeployViaCopy, which
made go test's vet step fail the whole package.

Add smartlink tests for an existing target, concurrent links to the same
destination, and the Link refactor's handling of directories and
symlinks.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* Skip per-file metadata calls when linking

link() checked each file with IsDir, IsSymlink, MkdirUnlessExists on its
parent and TargetExists before linking it, plus another IsDir inside the
Windows linkFile. Each is a file open on Windows. Instead:

- take the entry type from the directory listing (os.ReadDir already
  has it), so only symlinks need a Stat;
- rely on the parent directory existing, since the recursion creates it
  (Link creates it for its single entry);
- drop the TargetExists pre-check and treat the link call's "already
  exists" error as the skip, which already covered concurrent links;
- check for a directory only on the symlink-to-directory path, which is
  the only one that can pass one to linkFile.

A regular file now costs one syscall: the link itself.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Bump github.com/klauspost/compress (indirect, via mholt/archives) from
v1.18.0 to v1.18.7, the first patched release for GHSA-259r-337f-4rfw
(integer overflow and out-of-bounds write in s2). Re-vendored.

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Support dependencies for encrypted private ingredients

`state publish --build` can now declare and edit the dependencies of an
encrypted private ingredient:

- Read [project].dependencies from pyproject.toml and publish them as
  runtime dependencies on language/python packages, with names normalized
  by the inventory API. Version ranges carry over (~= and ==X.* are
  expanded). OS and Python version markers become conditions
  (kernel/Windows|Linux|Darwin, language/python); unsupported markers are
  skipped with a notice. The wheel's METADATA records them as
  Requires-Dist.
- Read the --meta file before building, so a meta file's name, version
  and description feed the wheel (flags > meta file > pyproject.toml).
- Add `state export ingredient-meta <namespace>/<name>[@Version]`, which
  writes an existing ingredient's meta file for use with --edit --meta.

Fix --edit along the way: keep the meta file's dependencies and authors
instead of overwriting them with the previous revision's, keep inherited
dependency types and conditions, add the private-builder dependency only
once, and reject author/description changes on an existing version up
front instead of failing in the API. InventoryRequirementsToString no
longer panics on requirements without a version.

Add a readme walking through the publish and edit workflow.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* Address review: keep inherited deps, edit the meta file's version

- When every pyproject.toml dependency is skipped, keep the dependencies
  already in place (on --edit, those inherited from the previous
  revision) instead of replacing them with an empty list.
- Let a meta file's version select the version to publish or edit, like
  --version (which still takes precedence). An exported meta file passed
  back with `state publish --edit --meta` without --build was editing
  the latest version instead of the exported one.

Add tests for both, for the up-front author/description rejection on
--edit, for InventoryRequirementsToString with requirements missing a
version or comparator, and for inherited conditions that carry
requirements.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@icanhasmath icanhasmath mentioned this pull request Oct 22, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

7 participants