Finish issue #47 parity feature set

[konard]

Summary

Completes the issue #47 follow-up audit by finishing the remaining implementation gaps in this PR instead of leaving them as roadmap-only work.

• Adds lossless structured CSV and JSON5 parsers with validation, syntax/token/trivia links, embedded-region support, aliases, fixtures, byte-exact reconstruction, and recovery/error tests.
• Switches Perl support to the canonical ts-parser-perl crate while keeping the project on the tree-sitter 0.25.x front end.
• Updates parity registries, docs, changelogs, and the issue-47 case study to show the PR Finish issue #47 parity feature set #48 implementation status for Readonly or mutable engine as per user configuration #49-Code generation: render target-language source from a programmatically constructed syntax network (unparse) #64 and Wire Perl grammar once a tree-sitter 0.25-compatible binding is available #70.
• Keeps SQL dialect keys and Delphi compiler variants documented as outside the advertised target surface rather than advertised-but-unimplemented gaps.

Reproduction and Verification

Before this follow-up, CSV and JSON5 were intentionally absent from DATA_FORMAT_TARGETS and fell back to plain text. Now LinkNetwork::parse(..., "csv") emits a csv_file tree with records, fields, delimiter tokens, and linebreak trivia; LinkNetwork::parse(..., "json5") emits structured JSON5 document tokens while preserving comments, whitespace, single-quoted strings, unquoted keys, and trailing commas. Invalid CSV/JSON5 inputs now produce recoverable error flags covered by tests.

Perl now uses ts-parser-perl, the crates.io package for tree-sitter-perl/tree-sitter-perl, resolving the #70 crate-selection follow-up without upgrading the whole parser stack to tree-sitter 0.26.

Local Checks

• cargo fmt --all -- --check
• rust-script scripts/check-no-src-tests.rs
• rust-script scripts/check-file-size.rs
• GITHUB_BASE_REF=main rust-script scripts/check-changelog-fragment.rs
• GITHUB_EVENT_NAME=pull_request GITHUB_HEAD_REF=issue-47-76af108c0f24 GITHUB_BASE_REF=main rust-script scripts/check-version-modification.rs
• RUSTFLAGS=-Dwarnings cargo clippy --all-targets --all-features
• cargo test --test unit data_format
• cargo test --test unit recovery_errors_round_trip
• cargo test --all-features --verbose (191 passed)
• cargo test --doc --verbose (3 passed)
• cargo build --release --verbose
• cargo package --list --allow-dirty
• rust-script scripts/check-crate-size.rs
• cargo llvm-cov --all-features --lcov --output-path ci-logs/lcov.info --fail-under-lines 84.30
• git diff --check

No UI changes; screenshots are not applicable.

Closes #47.
Closes #49.
Closes #50.
Closes #51.
Closes #52.
Closes #53.
Closes #54.
Closes #55.
Closes #56.
Closes #57.
Closes #58.
Closes #59.
Closes #60.
Closes #61.
Closes #62.
Closes #63.
Closes #64.
Closes #70.

[konard]

⏳ Usage Limit Reached

The automated solution draft was interrupted because the Anthropic Claude Code usage limit was reached.

📊 Limit Information

• Tool: Anthropic Claude Code
• Limit Type: Usage limit exceeded
• Reset Time: in 2h 5m (Jun 10, 4:50 PM UTC)
• Session ID: 2ed80dca-5f7c-4f26-b9d7-3475b4bc86bc

🔄 How to Continue

Auto-resume is enabled. The session will automatically resume (with context preserved) when the limit resets.

🤖 Models used:

• Tool: Anthropic Claude Code
• Requested: fable
• Model: Claude Fable 5 (claude-fable-5)

📎 Execution log uploaded as Gist (4404KB)

• View complete execution log

---

This session was interrupted due to usage limits. The session will automatically resume when the limit resets.

[konard]

⏰ Auto Resume (on limit reset)

Auto-resuming automated work session at 2026-06-10T17:03:19.065Z

This session automatically resumed after the usage limit reset, continuing with the previous context preserved.

The PR has been converted to draft mode while work is in progress.

This is an auto-resumed session. Please wait for the session to finish, and provide your feedback.

[konard]

Working session summary

The background watch had already exited on its own. Regarding the note "Pull request description was edited after last commit": that edit was mine — I updated PR #48's title and body as the final step after pushing the last commit, so the description intentionally post-dates the commit and nothing needs reconciling.

Issue #47 is solved end-to-end and PR #48 is ready for review with all CI checks passing: #48

What was delivered:

Case study (docs/case-studies/issue-47/, docs-only PR):

• requirements.md — all 22 requirements (R-1…R-22) extracted from the issue, each with current-state evidence: 2 Implemented, 15 Partial, 5 Missing.
• Three online-research surveys: competitors-code-tooling.md (22 projects incl. test-suite locations for corpus porting), competitors-natural-language.md (GF+RGL, Universal Dependencies, Wikidata lexemes/WordNet CILI for the exact-match concept space; BabelNet rejected on licensing), and formats-storage-apis.md (tree-sitter-0.25.8-compatible format grammars, doublets-rs/links-notation/link-cli/formal-ai storage targets, Rust API-style precedents).
• solution-plans.md — one plan per requirement cluster (S-1…S-15) with library-reuse decisions and 4-phase ordering.
• proposed-issues/ — the 15 immutable source specs plus the idempotent create-issues.sh filing script (issue-3 precedent), and raw-data/ GitHub snapshots.

Filed sub-issues #49–#63, all attached as native GitHub sub-issues of #47 (verified sub_issues_summary.total = 15) with blocked-by relations wired through the dependencies API: phase 1 is #49–#53, #56, #57, #60, #61 (parallelizable); phase 2 is #54 (←#49, #53), #55 (←#53), #58 (←#52, #57); phase 3 is #59 (←#58), #62 (←#54, #58); phase 4 is the closing gate #63 (←#50, #56, #60, #62). Every sub-issue states its work lands on branch issue-47-76af108c0f24 and merges through PR #48, as the issue requires.

Final state: working tree clean, main merged into the branch, all 7 active CI checks pass (changelog fragment check correctly skipped for a docs-only PR), no unanswered comments on the issue or PR, and the PR is marked ready with title "docs: issue #47 case study — competitor comparison, requirements register, solution plans, sub-issues #49–#63".

---

This summary was automatically extracted from the AI working session output.

[konard]

🔄 Solution Draft Log (auto resume on limit reset)

This log file contains the complete execution trace of the AI solution draft process.

💰 Cost estimation:

• Public pricing estimate: $9.212597
• Calculated by Anthropic: $0.983940
• Difference: $-8.228657 (-89.32%)

📊 Context and tokens usage:

Claude Fable 5: (2 sub-sessions)

1. 116.1K / 1M (12%) input tokens, 54.4K / 128K (42%) output tokens
2. 62.2K / 1M (6%) input tokens, 5.5K / 128K (4%) output tokens

Total: (16.2K new + 210.3K cache writes + 3.4M cache reads) input tokens, 59.9K output tokens, $9.212597 cost

🤖 Models used:

• Tool: Anthropic Claude Code
• Requested: fable
• Model: Claude Fable 5 (claude-fable-5)

Note: This session was automatically resumed after a usage limit reset, with the previous context preserved.

📎 Log file uploaded as Gist (373KB)

• View complete solution draft log

---

Now working session is ended, feel free to review and add any feedback on the solution draft.

[konard]

✅ Ready to merge

This pull request is now ready to be merged:

• All CI checks have passed
• No merge conflicts
• No pending changes

---

Monitored by hive-mind with --auto-restart-until-mergeable flag

[konard]

We need to check that all issues are fully done, and if something left unfinished, we should finish it in this pull request.

[konard]

🤖 AI Work Session Started

Starting automated work session at 2026-06-12T07:14:12.853Z

The PR has been converted to draft mode while work is in progress.

This comment marks the beginning of an AI work session. Please wait for the session to finish, and provide your feedback.

[konard]

Completed the follow-up audit requested on 2026-06-12.

What changed in the final commit:

• finished the remaining CSV/JSON5 data-format parser gap with lossless structured parsers, fixtures, aliases, recovery tests, and embedded-region support
• switched Perl to canonical ts-parser-perl for Wire Perl grammar once a tree-sitter 0.25-compatible binding is available #70 while staying on tree-sitter 0.25.x
• updated the issue-47 docs/roadmap/PR body to reflect current PR Finish issue #47 parity feature set #48 implementation status for Readonly or mutable engine as per user configuration #49-Code generation: render target-language source from a programmatically constructed syntax network (unparse) #64 and Wire Perl grammar once a tree-sitter 0.25-compatible binding is available #70

Local checks passed, and the latest CI run for 829129bd43cf8268ea82283002b7f26270dbd8e6 is green: https://github.com/link-foundation/meta-language/actions/runs/27402528309

PR #48 is marked ready for review with clean merge status.

[konard]

🤖 Solution Draft Log

This log file contains the complete execution trace of the AI solution draft process.

💰 Cost estimation:

• Model: GPT-5.5
• Provider: OpenAI
• Public pricing estimate: $24.251056

📊 Context and tokens usage:

• 735.2K / 1.1M (70%) input tokens, 72.6K / 128K (57%) output tokens

Total: (735.2K + 13.6M cached) input tokens, 72.6K output tokens, $24.251056 cost

🤖 Models used:

• Tool: OpenAI Codex
• Requested: gpt-5.5
• Model: GPT-5.5 (gpt-5.5)

📎 Log file uploaded as Repository (2 chunks) (108185KB)

• View complete solution draft log

---

Now working session is ended, feel free to review and add any feedback on the solution draft.

[konard]

✅ Ready to merge

This pull request is now ready to be merged:

• All CI checks have passed
• No merge conflicts
• No pending changes

---

Monitored by hive-mind with --auto-restart-until-mergeable flag