omp
621679cee5
fix: corrige bugs críticos de índice de páginas e cache de OCR
...
- C-1: corrige índice de páginas OCR'd (retorna page_number em vez de position)
- C-2: adiciona tratamento para falhas de OCR no cache (não distingue erro de não processado)
- Integra paralelização do OCR via subprocessos (parallel.py)
- Indexador usa extração paralela quando disponível
- Melhora relatório de progresso com contagem de páginas processadas
2026-11-01 20:01:28 -03:00
omp
b3f8019e01
feat: report indexing progress with page counts in status panel
2026-11-01 19:32:33 -03:00
omp
8f79993646
fix(backend): detect corrupted PDF text layer by control character density
2026-11-01 19:30:07 -03:00
omp
b25bf8ff7d
feat: add Docker Compose, environment template and README
2026-11-01 19:25:55 -03:00
omp
7f4d4bde66
feat(frontend): add React chat UI with SSE parsing and source accordion
2026-11-01 19:25:35 -03:00
omp
7a0396422f
feat(backend): add FastAPI routes with SSE chat and reindex guard
2026-11-01 19:24:20 -03:00
omp
99bc46f053
feat(backend): add chat service with evidence gating and citation validation
2026-11-01 19:19:45 -03:00
omp
9be09921bc
feat(backend): add LLM provider adapters for Ollama, OpenAI, Groq and Anthropic
2026-11-01 19:18:07 -03:00
omp
1748b561bd
feat(backend): add incremental indexer with manifest and partial states
2026-11-01 19:17:07 -03:00
omp
abf7d802ab
feat(backend): add embeddings and ChromaDB vector store
2026-11-01 19:16:40 -03:00
omp
6dc7f634ea
feat(backend): add biblical parser with book/chapter/verse reconstruction
2026-11-01 19:15:22 -03:00
omp
b75b746a40
feat(backend): add PDF extraction with OCR fallback and cache
2026-11-01 19:14:06 -03:00
omp
1b860dc599
chore: add gitignore
2026-11-01 19:12:42 -03:00