refactor: 0.6.x cleanup batch — cross-surface policy single-sourcing (architecture review F4/F5/F7) - #52
Merged
Merged
Conversation
…(architecture review F4/F5/F7) Move the last policy duplications into models (BOOLEAN_OPTIONS, recommended_first, RATIONALE_HEADERS, PROGRESS_STATUS_ICONS), delete the bridge's shadow of CONFIRM_DEFAULT_OPTIONS so the 0.5.0 single-sourcing claim is true, type the MCP field schema's object arrays with a grammar-coverage drift test, and fix the stale ATTUNE_KEYBOARD_MODE docstrings. Output byte-identical — pinned by the characterization suite plus the new tests/test_single_sourcing.py. Deviation from the review plan, recorded: the _EXTRAS_PARSERS registry was dropped. form_from_dict's parsers are interdependent (confirm reassigns options; ranking/assumption override suggested) — a uniform-signature registry would hide that dataflow behind a protocol, which is the abstraction the house philosophy prohibits. The completeness risk it was meant to pin is covered by Batch C's grammar-completeness test instead. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
silversurfer562
force-pushed
the
claude/forms-cleanup-batch
branch
from
August 20, 2026 13:13
e47d35b to
614cb2b
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Batch A of the 2026-08-20 dynamic-forms architecture review — the successor to the 0.5.0 cleanup batches (#34/#36). Output is byte-identical; the characterization suite plus a new pin file prove it.
Changes
models(F4b):BOOLEAN_OPTIONS(was defined independently in bridge and widget),recommended_first()(was implemented three times — widget, markdown surface, inline into_ask_user_format),RATIONALE_HEADERSandPROGRESS_STATUS_ICONS(each surface carried its own copy guarded only by a "matches the widget" comment). Newtests/test_single_sourcing.pypins each surface to the shared source.bridge._CONFIRM_DEFAULT_OPTIONSdeleted (F4a): the bridge now consumesmodels.CONFIRM_DEFAULT_OPTIONS— the 0.5.0 changelog's single-sourcing claim is now true._field_schemahygiene (F5): object-array extras typed (progress_items,triage_items,consequences,assumptions) plus a drift test — everyQuestionTypevalue must appear in the schema's type enum and everyFormQuestionfield in its properties. The prose description stays hand-written on purpose (the review explicitly rejected generating the schema from models).keyboard_mode_enabledand the package overview now documentATTUNE_FORMS_KEYBOARD_MODE(preferred) withATTUNE_KEYBOARD_MODEas legacy fallback, matching the code.Deviations from the approved plan (recorded)
_EXTRAS_PARSERSregistry dropped:form_from_dict's parsers are interdependent (confirm reassignsoptions; ranking/assumption overridesuggested) — a uniform-signature registry would hide that dataflow behind a protocol, exactly the abstraction the house philosophy prohibits. The completeness risk it was meant to pin lands in Batch C's grammar-completeness test instead.response_iduuid suffix (F8): already fixed on main (discovery-sweep finding, same day) — no change needed.Verification
Full suite: 689 passed (existing characterization pins untouched and green).
🤖 Generated with Claude Code