feat: standard names pipeline, DD tools improvement, dot-notation support - #32
Merged
Merged
Conversation
Add LLM-backed standard name composition pipeline: - Source extraction plugins for DD paths and facility signals - Compose workers with structured LLM output via Pydantic models - Cross-model review, benchmarking, and publish commands - CLI interface (imas-codex sn compose/review/benchmark/publish) - Pipeline state machine with graph-backed persistence Extract StandardName to dedicated cross-domain schema: - New imas_codex/schemas/standard_name.yaml bridges DD and facility - StandardName, StandardNameSource enum, StandardNameReviewStatus enum - HAS_STANDARD_NAME relationship declared for schema compliance - Both IMASNode and FacilitySignal can reference StandardName
Add implementation plans for standard names pipeline phases: - Plans 09-14: LLM compose, pipeline fixes, publish validation, rich compose, catalog import, MCP tools benchmark - Research documents: implementation review, archived plans - Superseded plans preserved for reference
…acked structure Improve 4 poorest-performing DD tools based on A/B testing: - search_dd_paths: add semantic cluster results alongside path matches - check_dd_paths: add fuzzy typo suggestions via PathFuzzyMatcher - get_dd_overview: physics domain aggregation, lifecycle stats, IDS descriptions - get_ids_structure: graph-backed COCOS fields, semantic clusters, identifiers, coordinate arrays — 4KB vs prod's 20KB for equivalent information Rename internal _imas_ tool methods to _dd_ convention for DD-specific functions. Recover ~470 lines of search intelligence lost during rename commit. Add dot-notation normalization with natural-language safety guard: - _looks_like_path() rejects NL text (e.g., 'B0.', 'eV.s', version strings) - Strips annotations before checking (handles 'equilibrium.time_slice(itime)') - Space-separated multi-path input correctly splits before normalizing Fix PathFuzzyMatcher: swap constructor arguments from (paths, ids_names) to correct order (ids_names, paths) — was silently treating 20K paths as IDS names and 87 IDS names as paths. Add format_cocos_fields_report and format_dd_changelog_report formatters. Fix server.py method name mismatches for cocos and error field tools.
Comprehensive test coverage for path normalization safety: - TestLooksLikePath: validates guard rejects NL text, versions, units - TestNormalizeImasPathDotNotation: dot-to-slash conversion cases - TestNormalizeImasPathAnnotations: bracket/parenthesis stripping - TestNormalizeImasPathEdgeCases: empty, no-dots, already-slash paths - TestNormalizePaths: space-separated, comma-separated, JSON array, mixed notation inputs
Align git sync discipline with fork-based development workflow. Clarify push targets: origin (fork) for daily work, upstream via PR only.
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
This PR brings three major feature areas validated through comprehensive A/B testing against production.
Standard Names Pipeline
imas-codex sn compose/review/benchmark/publishStandardNameschema bridging DD and facility signalsHAS_STANDARD_NAMErelationship for schema complianceDD Tools Improvement (A/B tested against production)
Based on systematic A/B testing of all 17 DD tools vs 11 production tools:
_imas_tool methods to_dd_conventionformat_cocos_fields_reportandformat_dd_changelog_reportformattersDot-Notation Support with Natural Language Safety
normalize_imas_path()convertscore_profiles.profiles_1d.electrons.temperatureto slash notation_looks_like_path()guard rejects natural language (e.g.,B0.,eV.s, version strings)equilibrium.time_slice(itime).profiles_1d.psiA/B Test Results
DD server average: 8.6/10 vs production: 5.5/10 (unique value to frontend LLMs)
Commits