tobi/grammar-inference-engine
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
18 commits 2 branches 0 tags 5.6 MiB
Commit graph

18 commits

This branch
This branch
All branches
Author SHA1 Message Date
tobjend
c8a49f0149 scale PNG logo to 150% 2026-07-01 11:21:02 +02:00
tobjend
923ecfd231 half the gif display size 2026-07-01 11:20:24 +02:00
tobjend
9d71763b27 restore original dervish.gif below explanation; add new logo left-aligned, smaller 2026-07-01 11:18:13 +02:00
tobjend
64c3313da9 replace logo with dervis_logo.png; add to SHOWCASE.md and blog_post.md 2026-07-01 11:16:21 +02:00
tobjend
67369f31a8 remove context-cost chart (too blatant) 2026-07-01 11:06:39 +02:00
tobjend
17b5c271ec deduplicate: replace detailed Real-world Results with summary table linking to SHOWCASE.md 2026-07-01 11:05:26 +02:00
tobjend
b05c3ee116 rename to Dervish MCP; expand description with token-savings framing; add xkcd-style bar charts; link papers to actual URLs 2026-07-01 11:05:03 +02:00
tobjend
6d1c033267 remove Terraform showcase (everything-optional grammar isn't useful); fix GHA scope claim 2026-07-01 10:42:08 +02:00
tobjend
d7477344a6 move format-specific adapters to examples/, purge format-specific MCP tools 2026-07-01 10:36:14 +02:00
tobjend
25d844d1f9 purge Portainer references, format-specific tools, and Domain Adapters section; make showcases concrete with extracted types 2026-07-01 10:36:04 +02:00
tobjend
097dfc9954 Rename to Dervish, add animated logo to README 2026-07-01 10:19:08 +02:00
tobjend
a8a8bddb37 Move MCP server to top of README — it's the primary interface 
Restructure: MCP Server first (with agent workflow example), then
Why grammar inference / showcases, then Quick Start, then details.
This matches how users actually interact with the project: via MCP tools.
 2026-07-01 10:18:10 +02:00
tobjend
9f5bde22d5 Remove bugs section (implementation bugs, not paper bugs), remove Docker Compose (private data), add Portainer templates, fix geerlingguy claim precision 
Blog post: remove 'The bugs we found' section (all 4 bugs were from our implementation, not the paper algorithms). Replace company data references in MCP section with Galaxy example. Update ensemble dynamics table with public datasets.

README: replace Docker Compose with Portainer templates in 'Why grammar inference?' table, Real-world Results, and Domain Adapters.

SHOWCASE: replace Docker Compose with Portainer templates.

All claims verified: no public documentation of geerlingguy module ordering convention exists.
 2026-07-01 10:15:22 +02:00
tobjend
547376894c Update README and SHOWCASE with real-world dataset evaluations 
README:
- Replace outdated company benchmarks with public showcases
- Add Algorithm Selection Guide
- Add 'When each algorithm wins' table
- Add 'Why grammar inference?' table with value prop for LLMs
- Add 'What doesn't work' section documenting failed approaches
- Update all domain adapter examples with public results
- Clean up outdated references (companyweb roles, hashistack terraform)

SHOWCASE:
- Add Helm (kube-prometheus-stack) with iDRegEx minimal core
- Add Docker Compose per-project patterns
- Add GitHub Actions cross-project Go lint pattern
- Add Terraform modules with vocabulary analysis
- Add 'What doesn't work' section
- Explain WHY each dataset helps an LLM
 2026-07-01 10:04:10 +02:00
tobjend
0e2aec582b Grammar inference engine: CRX + iDRegEx ensemble with MDL scoring, MCP server, showcase, and blog post 
- Ensemble inference (infer_ensemble) runs both CRX and iDRegEx, picks best by MDL
- CRX: CRX algorithm for wide coverage (accepts all sequences, large vocabulary)
- iDRegEx: iDRegEx for minimal core grammar (tightest common pattern)
- MDL scoring: fixed model_cost to count alphabet symbol occurrences, fixed dispatch order in _count_words_fast
- Fixed _match_tokens: rewritten as _match_possible with proper backtracking
- Fixed _parse_parts disjunction: children use _parse_flat_symbol to avoid dot-splitting
- MCP server: infer_best_grammar and infer_grammar tools
- Added prefer parameter (crx/idregex) to skip ensemble
- 28 passing tests
- SHOWCASE.md with Geerlingguy Galaxy demonstration
- blog_post.md with full technical deep-dive
 2026-07-01 09:51:41 +02:00
tobjend
a1567bffbe Add bin/mcp-server wrapper script for robust path resolution 
Avoids module shadowing when CWD contains another bex package.
 2026-07-01 08:06:17 +02:00
tobjend
adc52c99ec Add MCP server: grammar inference via FastMCP 
- bex/mcp_server.py: FastMCP server with 3 tools:
  * infer_grammar(sequences, method='crx'|'idregex')
  * infer_yaml_grammar(yaml_dir, pattern, method)
  * infer_ansible_role_grammar(roles_dir)
- pyproject.toml: add bex-mcp console_scripts entry point
 2026-07-01 08:03:10 +02:00
tobjend
7c00c6713d Initial commit: BEX-based grammar inference engine 
- CRX: direct CHARE inference (Algorithm 7, TODS 2010)
- iDRegEx: k-ORE inference (Algorithm 4, arXiv 2010)
- RWR₀: SORE repair (Algorithm 6, TODS 2010)
- rwr²: k-ORE extraction (Algorithm 3, arXiv 2010)
- SOA, k-OA, iKoa, 2T-INF, Baum-Welch
- Ansible role grammar adapter
- Generic YAML key-path converter
- 28 tests, all passing
 2026-07-01 08:01:16 +02:00