PyCForge Phase 10 v0.10.0 Project Handoff
==========================================

Status
------

Phase 10 is complete and promoted as PyCForge 0.10.0. The release delivers the
bounded support-template infrastructure required by Architecture Revision 3.1
and its active Revision 3.2 addendum. The cumulative supported Python boundary
is unchanged from Phase 9.

The sealed Phase 9 release was not modified. Its exact rollback identities are:

* archive: pycforge_phase_9_v0_9_0.tar.gz
* archive SHA-256: 68a5bbe443513d5a40a009be8e55ca9ec513805a4dc6f8c9d5e08bdd6a4afcff
* sealed tree SHA-256: bfbb13eb764b02a6b8fb2c4ff1eb12f4249976bb1919aa85a383d6e24b4079e8

The supplied Phase 10 opening checkpoint was authenticated before development:

* opening archive SHA-256: 75405ec38815be29b287b838ba0ff866dfe6ac1dc8cd6124551b2637a1065904
* opening tree SHA-256: f9ff0610c084610d917e4a11a68b83a95ab6aea8746276f3e6771c3b64e5fd18
* opening tests: 143 passed

Roadmap custody
---------------

The full authoritative roadmap and addendum remain packaged:

* Architecture Revision 3.1 SHA-256:
  d05963984105466ae7df4c0e25941865fb0e55181059e3c95e3fde42556fd8e3
* Revision 3.2 addendum SHA-256:
  93b228528a10895fdbdce7b6934d422e7eddd1a79b409bdbc15ea6677e5ab8c6

Historical whole-tree validators remain scoped to their sealed historical
trees. Later cumulative trees prove compatibility through the complete
regression suite, predecessor identities, and the current-phase validator.

Phase 10 delivery
-----------------

The accepted entry-gate decisions are exact, genuine, and non-promoting:

* pycf.i64.floor_div@1.0.0
* pycf.i64.floor_mod@1.0.0

They document prospective bounded-int64 Python/C rounding obligations,
caller-proved divisor and overflow preconditions, scalar-by-value ownership,
no allocation or cleanup, no runtime failure channel, and eligibility only in a
separate numeric-semantics mini-phase no earlier than Phase 14. They authorize
infrastructure and fixtures only. Python floor division and modulo remain
unsupported.

The release provides:

* frozen project-owned helper registry phase10-support-templates-v0.10;
* exact interface pycforge-helper/1 and exact semantic-version references;
* complete dependency closure, deduplication, deterministic topological order,
  cycle and missing-dependency diagnostics, and unused-helper exclusion;
* Target C Source Contract, interface, ownership/lifetime, failure, provenance,
  static-linkage, identifier, structured-C-IR, and rendered-text validation;
* structured C IR factories only, with no raw-text or user-template path;
* deterministic manifests, registry/asset/closure fingerprints, conversion
  summaries, Full-trace evidence, and PYC33xx helper diagnostics;
* cancellation without a partial helper plan, artifact, or generated source;
* exact helper golden sources and source/installed helper audits;
* the completed Python-first, hidden-read-only-C, non-modal-progress workspace
  gate from the opening checkpoint.

Current promoted RulePlans declare no helper requirements. The final helper
manifest is empty for the existing supported subset, and representative
generated C remains byte-identical to the opening checkpoint.

Schema identities
-----------------

Phase 10 publishes generated-c/0.10 artifacts,
pycforge.conversion-summary/0.10 summaries,
pycforge.decision-trace/0.10 traces, and result serialization 0.3. With the
current empty helper manifest, the inner structured representation remains
c-ir/0.9. Actual helper assembly advances that inner representation to
c-ir/0.10.

Release evidence
----------------

* tests: 154 passed (143 opening-checkpoint regressions + 11 Phase 10 tests)
* audits: architecture, rules, helpers, determinism, and transition passed
* actual widget smoke: PyQt5 5.15.11 / Qt 5.15.14, offscreen, passed
* wheel: pycforge-0.10.0-py3-none-any.whl, 106869 bytes
* wheel SHA-256: 042285b57d93ede6443a6ae700283f639ac3957e7ebececd689f2ad25e87b5e7
* two fixed-epoch wheel builds: byte-identical
* isolated wheel install, conversion, schema checks, and installed helper audit:
  passed

Generated C and helper sources were validated structurally and textually only.
They were not compiled, linked, loaded, or executed.

Validation
----------

Run from the release root:

    python -m unittest discover -s tests
    python tools/validate_phase10.py
    python -m pycforge audit architecture
    python -m pycforge audit rules
    python -m pycforge audit helpers
    python -m pycforge audit determinism
    python -m pycforge audit transition --phase phase_10

Next objective
--------------

Phase 11 bounded containers has not started. Before any container form is
implemented, approve its representation and failure policy. Each admitted form
must define capacity, aliasing, ownership/lifetime, mutability or immutability,
ordering, indexing and negative-index behavior, bounds and allocation failure,
cleanup, element constraints, and exact helper requirements. Preserve the
single deterministic translation-unit architecture until a genuine boundary
requires more. Do not introduce broad import/library semantics, a Python
runtime, a broad helper library, dynamic resizing, or multi-file architecture
by convenience.
