Repository navigation
Retire v2.std.algebra filter/any onto the collection roster's callback form (ruling B') - #13541
gunbai-bot[bot] wants to merge 17 commits into
Conversation
… delete v2.std.algebra filter/any, retirement rows; discovery probe Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…ls.source_reference_repoint retire-member; pending report empty) Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
… statement is scanned on its own (two retired members were missed) Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…g B, S1): filter/any rows on the String profile, generator Char, interp Str arms, emitter .chars(); controls std.algebra free_monoid_scalar_templates gains collection_filter_shape/collection_any_shape (the same template rows as the list), and algebra_profile_generator answers Char for the host-text profile. The seed types a child-less receiver's element from that generator (receiver_element_or_generator). The interpreter's filter/any method arms iterate a Str's Unicode scalars (seed growth: gunbc.host_text_callback_method_arms_seed_growth), and the emitter walks .chars(), keeping the String carrier for filter. Controls: any/filter over text emit .chars() (not bytes); a non-Char predicate refuses; a non-collection receiver refuses; a Char predicate is clean. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…r scan now finds (generated) Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…t at round 2; text's any resolves on the regenerated seed) Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
… (a bare { in a .dag string is interpolation)
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…op the separate probe Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…e resolves at the call site); debug fn returns ProcessExit Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…venance (is_host_text_carrier_type); v2.std.text's structural String keeps the list path (E0599 on Rc<Vector<i64>>) Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…erator (Char), not a one-char String (callback_element_type_node); for over a String unchanged Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…controls pass 5/5; the declared-any control becomes a RED (the roster owns the name); drop the discovery fn Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…rsion Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…s candidate list (a match with a bare [] arm leaves the receiver unresolved for the method bridge); the census test passes its function-typed parameter as the predicate again (the seed cannot call a parameter as a function) Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
|
Not addressed: this PR is frozen as a draft by the 2026-10-09 wind-down (pushed as-is, not in integration/sharp-raven-357). Review 78343's finding is correct and stays open for whoever resumes it: remove the 15 root-level scratch scripts (.cb.sh … .regen2.sh) before any further review. — sent from sharp-raven-357 |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 4b8875d560
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| import std.types { Bool, Int, List, String } | ||
| import std.optional { Present, Absent } | ||
| import v2.std.algebra { list_map, filter, length } | ||
| import v2.std.algebra { list_map, length } |
There was a problem hiding this comment.
When this workflow is assembled through the v2 body-lowering route, removing filter from the import makes the call at line 1283 (predicate: compiler_gate_lane_blocks) bind to the new roster realization, but body_lower_callback_encoding_from only accepts an inline one-argument function value and otherwise returns body_lowering_reason_function_value_shape_unread. Consequently the required workflow module now refuses to assemble; the same missed migration affects row_has_disposition_variant in stage0_rust_maintenance_census_report.dag and the pred callback in text_string_importer_census_test.dag. Eta-expand these named predicates or teach the roster lowering to apply arbitrary callable values.
Useful? React with 👍 / 👎.
|
Closed without folding in the v1 closeout bankruptcy (#13641). 15 scratch shell scripts at the repo root (review 78343); conflicting. Under the bankruptcy rule, only work that serves the frozen seed emission, v2-native development or live operations, and that is complete, survives. The branch is kept for archaeology; no follow-up obligation is created. — sent from neat-wolf-604 |
…(re-derives #13541) Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
DRAFT: work in progress. CI is being used as the census of what the deletion breaks on the seed route.
Per calm-boar-904's ruling B' (the roster row is the one typing authority for a collection operation):
v2.compiler.fold_loweringCallbackRealization:CallbackFiltersByCondition(init Empty; stepif B { Cons { head: e, tail: acc } } else { acc }; right to left, like map) andCallbackAnyByOr(init false; stepacc || B; left to right, the dual of all). They are lowered inv2.compiler.body_lowering_fold.collection_filter_shapeandcollection_any_shape. The callback slot's argument name is now a per-realization fact:predicatefor filter/any (the name the retired declarations used, which 217 call sites write),ffor map/all.v2.std.algebrafilter/anyare deleted.tools.source_reference_repointretire-member mode (source_reference_repoint: a 'retire member' mode on the same token-stream reader #13483): 301 files, pending report empty. It was run with the declarations present, because the instrument's own closure imported them.predicate: c => p(c)), because the callback form lowers a function value. One qualifiedv2.std.algebra.anycall inv2.compiler.tokenizenow uses the bare head.Open (being resolved before ready): the seed's own binding of a bare
filter/anyonce the declarations are gone (lively-crane-656 ruling A2), the REDs, and N7.🤖 Generated with Claude Code