Commit Graph

7 Commits

Author SHA1 Message Date
c9a777b5fb fix deploy 2026-03-30 18:30:36 +02:00
775cd1aa47 refactor: flatten shell/ dir and add ask screenshot to README 2026-03-29 20:56:50 +02:00
ed5fa6756e refactor: reorganize project structure and fix broken references
- Move scripts to scripts/ directory (roda.sh, prepara_db.py, etc.)
- Move shell config to shell/ directory (Caddyfile, auth.py, haloy.yml)
- Move basedosdados.duckdb to data/ directory
- Update Dockerfile and start.sh with new file paths
- Update README.md with correct script paths
- Remove Python ask.py (replaced by Rust binary in ask/ask)
- Add Rust source files (schema_filter.rs, sql_generator.rs, table_selector.rs)
- Remove sentence-transformer dependencies from ask
- Move docs and context artifacts to their directories
2026-03-29 20:46:27 +02:00
3788e2cc81 fix: use Python ask.py instead of Rust binary to avoid compilation 2026-03-28 15:46:48 +01:00
a6509d8b30 Add logging to ask app: save questions, SQLs, success/error status, and timestamps to logs/log.json 2026-03-28 12:17:34 +01:00
e1c2377343 feat(ask): add text wrapping for wide table columns
- Implement wrap_text function to handle long cell content
- Auto-wrap table columns when content exceeds available width
- Preserve original table rendering for fits-all cases
- Remove sample_datasets project (no longer needed)
- Update .gitignore to use wildcard for target dirs
2026-03-28 11:59:02 +01:00
b5d84e3556 feat: add LLM SQL query assistant and dataset sampler
- ask.py: Python script to query Base dos Dados via natural language using Gemini,
  generates and executes DuckDB SQL from Portuguese questions
- ask/ (Rust): CLI companion for the SQL query assistant with system prompt
- sample_datasets.py: samples parquet files from S3 into a local DuckDB for exploration
- sample_datasets/ (Rust): CLI for dataset sampling
- context/: LLM context bundle (schemas, join keys, file tree) for query generation
2026-03-28 11:23:51 +01:00