Files
hakorune/docs/development/current/main/phases/phase-29aq/README.md

64 lines
2.9 KiB
Markdown
Raw Normal View History

---
Status: Active
Scope: stdlib trim/parse loop inventory (docs-first)
Related:
- docs/development/current/main/design/coreplan-migration-roadmap-ssot.md
- docs/development/current/main/phases/phase-29ae/README.md
- docs/development/current/main/phases/phase-29ap/README.md
---
# Phase 29aq: stdlib trim/parse loop inventory (docs-first)
Goal: inventory loop shapes in json_native stdlib and map them to conservative
Plan/Composer subsets (or mark unsupported) before adding new subsets.
## Ok(None) / Freeze boundary (SSOT)
- Ok(None): non-matching shapes (default).
- Freeze: only when a shape is explicitly gated and contradictory; do not add
new Freeze cases in P0.
## Fixture / smoke naming (SSOT)
- Fixtures: `apps/tests/phase29aq_<area>_<func>_<shape>_min.hako`
- Smokes: `tools/smokes/v2/profiles/integration/joinir/phase29aq_<area>_<func>_<shape>_vm.sh`
- Gate: wire into `phase29ae_regression_pack_vm.sh` when a subset is added.
## Inventory (stdlib json_native)
### StringUtils (`apps/lib/json_native/utils/string.hako`)
| Function | Loop summary | Provisional bucket | Notes |
| --- | --- | --- | --- |
| `trim_start` | scan until non-whitespace, `break` | Pattern2Break | Subset already exists (Phase 29ap P6) |
| `trim_end` | reverse scan until non-whitespace, `break` | Pattern2Break | Subset already exists (Phase 29ap P6) |
| `index_of` | scan, early `return` on match | ScanWithInit | Needs return-in-loop handling |
| `last_index_of` | reverse scan, early `return` | ScanWithInit | Needs return-in-loop handling |
| `index_of_string` | scan substrings, early `return` | SplitScan | Candidate subset |
| `to_upper` | scan + accumulate | ScanWithInit | Candidate subset |
| `to_lower` | scan + accumulate | Pattern1 subset | Already migrated (Phase 29ap P2) |
| `join` | array join with `if i > 0` separator | Pattern1 subset | Already migrated (Phase 29ap P3) |
| `split` | scan + push segments | SplitScan | Candidate subset |
| `is_integer` | scan digits, early `return false` | Unsupported | Return-heavy; needs design |
| `parse_integer` | scan digits, `break` on invalid | Pattern2Break | Candidate subset |
### Lexer / Parser (out of scope for P0)
| File | Loop summary | Provisional bucket | Notes |
| --- | --- | --- | --- |
| `lexer/scanner.hako` | scan loops with nested loops (string/number) | Unsupported | Multiple nested scans |
| `lexer/tokenizer.hako` | tokenize loop with `break`/`continue` | Unsupported | Control-flow heavy |
| `parser/parser.hako` | `loop(true)` object/array parse | Unsupported | Structured parse loop |
| `utils/escape.hako` | escape/unescape scans | Unsupported | Nested scan/branching |
| `core/node.hako` | collection iteration loops | Unsupported | Not a parsing scan |
## Next (planned)
## Progress
- P1: Add stdlib subsets in priority order (index_of/last_index_of → parse_integer → split).
## Next (planned)
- P2: Extend stdlib scan subsets (candidate: index_of_string, to_upper).