chore: auto-commit (95 archivos)

- cmd/fn/doctor.go - cmd/fn/main.go - cpp/apps/primitives_gallery/playground/tables/CMakeLists.txt - cpp/apps/primitives_gallery/playground/tables/data_table.cpp - cpp/apps/primitives_gallery/playground/tables/data_table_logic.cpp - cpp/apps/primitives_gallery/playground/tables/data_table_logic.h - cpp/apps/primitives_gallery/playground/tables/self_test.cpp - cpp/apps/primitives_gallery/playground/tables/tql.cpp - cpp/apps/primitives_gallery/playground/tables/viz.cpp - cpp/apps/primitives_gallery/playground/tables/viz.h - ... Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-13 00:50:34 +02:00
parent a2bbf23374
commit e3c8979e8d
189 changed files with 18964 additions and 330 deletions
@@ -0,0 +1,90 @@
+---
+name: vault_audit
+kind: pipeline
+lang: bash
+domain: pipelines
+version: "1.0.0"
+purity: impure
+signature: "vault_audit(<vault_name> | --all) [--skip-profilers] [--dry-run-layout] -> void"
+description: "Pipeline completo de auditoria para uno o todos los vaults declarados: layout-ensure, index, profile (csv/pdf/md), dedupe, aggregate y doctor. Produce tabla resumen con estado por vault y codigo de salida 4 si hay warnings."
+tags: [vault, audit, pipeline, launcher, infra, bash]
+uses_functions:
+  - vault_layout_ensure_go_infra
+  - vault_inventory_scan_go_infra
+  - vault_index_open_go_infra
+  - vault_index_write_go_infra
+  - vault_csv_profile_py_datascience
+  - vault_pdf_extract_py_datascience
+  - vault_knowledge_parse_py_infra
+  - vault_dedupe_report_py_infra
+  - vault_aggregate_index_go_infra
+  - vault_doctor_go_infra
+uses_types: []
+returns: []
+returns_optional: false
+error_type: "error_go_core"
+imports: []
+params:
+  - name: vault_name
+    desc: "Nombre del vault a auditar (como aparece en registry.db tabla vaults). Usar --all para todos."
+  - name: --all
+    desc: "Audita todos los vaults declarados en registry.db. Mutuamente excluyente con vault_name."
+  - name: --skip-profilers
+    desc: "Omite el paso de profiling CSV/PDF/MD. Util para auditorias rapidas de inventario."
+  - name: --dry-run-layout
+    desc: "Pasa --dry-run a vault layout-ensure: calcula cambios sin tocar el disco."
+output: "Tabla de resumen por vault con status ok/warn. Codigo de salida 0=exito, 1=root no localizable, 4=uno o mas vaults con warnings."
+tested: false
+tests: []
+test_file_path: ""
+file_path: "bash/functions/pipelines/vault_audit.sh"
+---
+
+## Ejemplo
+
+```bash
+# Auditar un vault especifico
+FN_REGISTRY_ROOT=/home/lucas/fn_registry \
+  bash bash/functions/pipelines/vault_audit.sh turismo_spain
+
+# Auditar todos los vaults
+FN_REGISTRY_ROOT=/home/lucas/fn_registry \
+  bash bash/functions/pipelines/vault_audit.sh --all
+
+# Solo layout + index + aggregate (sin profilers, mas rapido)
+bash bash/functions/pipelines/vault_audit.sh turismo_spain --skip-profilers
+
+# Ver que haria layout-ensure sin tocar disco
+bash bash/functions/pipelines/vault_audit.sh turismo_spain --dry-run-layout
+
+# Equivalente via fn run (desde la raiz del registry)
+./fn run vault_audit_bash_pipelines turismo_spain
+```
+
+## Pasos del pipeline
+
+1. **layout-ensure** — `fn vault layout-ensure <name>` asegura `data/{raw,processed,exports}` y `knowledge/{...}`.
+2. **index** — `fn vault index <name>` escanea archivos y persiste en `vault_index.db`.
+3. **profile** — `fn vault profile <name>` llama `vault_profile_dispatch.py` para CSV/PDF/MD.
+4. **dedupe** — `fn vault dedupe <name>` detecta duplicados por sha256 (informacional, no fatal).
+5. **aggregate** — `fn vault aggregate` copia todo a `registry.db` tabla `vault_files` (una sola vez al final).
+6. **doctor** — `fn vault doctor` muestra estado de salud de todos los vaults.
+
+## Codigos de salida
+
+| Codigo | Significado |
+|--------|-------------|
+| 0 | Todos los vaults procesados sin errores |
+| 1 | FN_REGISTRY_ROOT no localizable o fn binary no encontrado |
+| 4 | Uno o mas vaults con warnings (layout o index fallaron) |
+
+## Variables de entorno
+
+- `FN_REGISTRY_ROOT` — raiz del registry (auto-detectada si no esta seteada).
+- `FN_BIN` — path al binario `fn` (default: `$FN_REGISTRY_ROOT/fn`).
+
+## Notas
+
+Requiere `sqlite3` en PATH para resolver la lista de vaults con `--all`.
+El paso de profile es non-fatal: errores en profilers individuales se reportan como warnings.
+El paso de dedupe es siempre informacional (no borra archivos).
@@ -0,0 +1,172 @@
+#!/usr/bin/env bash
+# vault_audit — Full audit pipeline for one or all declared vaults.
+# Runs: layout-ensure → index → profile → dedupe → aggregate → doctor
+#
+# Usage:
+#   vault_audit.sh <vault_name>
+#   vault_audit.sh --all
+#   vault_audit.sh <vault_name> --skip-profilers
+#   vault_audit.sh <vault_name> --dry-run-layout
+#   vault_audit.sh --all --skip-profilers
+set -euo pipefail
+
+# --- locate FN_REGISTRY_ROOT ---
+_find_registry_root() {
+    local dir
+    dir="$(pwd)"
+    while [[ "$dir" != "/" ]]; do
+        if [[ -f "$dir/registry.db" ]]; then
+            echo "$dir"
+            return 0
+        fi
+        dir="$(dirname "$dir")"
+    done
+    return 1
+}
+
+if [[ -n "${FN_REGISTRY_ROOT:-}" && -f "${FN_REGISTRY_ROOT}/registry.db" ]]; then
+    REGISTRY_ROOT="$FN_REGISTRY_ROOT"
+elif REGISTRY_ROOT="$(_find_registry_root 2>/dev/null)"; then
+    : # found
+else
+    echo "ERROR: Cannot locate registry.db. Set FN_REGISTRY_ROOT or run from registry root." >&2
+    exit 1
+fi
+
+FN_BIN="${FN_BIN:-${REGISTRY_ROOT}/fn}"
+if [[ ! -x "$FN_BIN" ]]; then
+    echo "ERROR: fn binary not found at $FN_BIN. Build with: CGO_ENABLED=1 go build -tags fts5 -o fn ./cmd/fn/" >&2
+    exit 1
+fi
+
+# --- parse args ---
+AUDIT_ALL=0
+SKIP_PROFILERS=0
+DRY_RUN_LAYOUT=0
+VAULT_NAMES=()
+START_TS=$(date +%s)
+
+while [[ $# -gt 0 ]]; do
+    case "$1" in
+        --all)           AUDIT_ALL=1 ;;
+        --skip-profilers) SKIP_PROFILERS=1 ;;
+        --dry-run-layout) DRY_RUN_LAYOUT=1 ;;
+        -*)
+            echo "ERROR: Unknown flag: $1" >&2
+            echo "Usage: vault_audit.sh <name> | --all [--skip-profilers] [--dry-run-layout]" >&2
+            exit 1
+            ;;
+        *)
+            VAULT_NAMES+=("$1")
+            ;;
+    esac
+    shift
+done
+
+if [[ $AUDIT_ALL -eq 0 && ${#VAULT_NAMES[@]} -eq 0 ]]; then
+    echo "Usage: vault_audit.sh <vault_name> | --all [--skip-profilers] [--dry-run-layout]" >&2
+    exit 1
+fi
+
+# --- resolve vault list ---
+if [[ $AUDIT_ALL -eq 1 ]]; then
+    mapfile -t VAULT_NAMES < <(
+        sqlite3 "${REGISTRY_ROOT}/registry.db" "SELECT name FROM vaults ORDER BY name;" 2>/dev/null || true
+    )
+    if [[ ${#VAULT_NAMES[@]} -eq 0 ]]; then
+        echo "No vaults registered in registry.db. Run 'fn index' first." >&2
+        exit 1
+    fi
+    echo "Found ${#VAULT_NAMES[@]} vault(s): ${VAULT_NAMES[*]}"
+fi
+
+# --- build fn vault flags ---
+LAYOUT_FLAGS=()
+if [[ $DRY_RUN_LAYOUT -eq 1 ]]; then
+    LAYOUT_FLAGS+=(--dry-run)
+fi
+
+# --- per-vault audit ---
+PASS_COUNT=0
+FAIL_COUNT=0
+declare -A VAULT_STATUS
+
+audit_one() {
+    local name="$1"
+    local vault_ok=1
+    echo ""
+    echo "=== vault: $name ==="
+
+    # Step 1: layout-ensure
+    echo "  [1/5] layout-ensure"
+    if ! "$FN_BIN" vault layout-ensure "$name" "${LAYOUT_FLAGS[@]}" 2>&1 | sed 's/^/    /'; then
+        echo "    WARN: layout-ensure failed (non-fatal)" >&2
+        vault_ok=0
+    fi
+
+    # Step 2: index
+    echo "  [2/5] index"
+    if ! "$FN_BIN" vault index "$name" 2>&1 | sed 's/^/    /'; then
+        echo "    ERROR: index failed" >&2
+        vault_ok=0
+    fi
+
+    # Step 3: profile
+    if [[ $SKIP_PROFILERS -eq 0 ]]; then
+        echo "  [3/5] profile"
+        if ! "$FN_BIN" vault profile "$name" 2>&1 | sed 's/^/    /'; then
+            echo "    WARN: profile had errors (non-fatal)" >&2
+        fi
+    else
+        echo "  [3/5] profile (skipped)"
+    fi
+
+    # Step 4: dedupe (informational, non-fatal)
+    echo "  [4/5] dedupe"
+    "$FN_BIN" vault dedupe "$name" 2>&1 | sed 's/^/    /' || true
+
+    # Step 5 deferred — aggregate runs once at the end
+    echo "  [5/5] aggregate (deferred)"
+
+    if [[ $vault_ok -eq 1 ]]; then
+        VAULT_STATUS["$name"]="ok"
+        PASS_COUNT=$((PASS_COUNT + 1))
+    else
+        VAULT_STATUS["$name"]="warn"
+        FAIL_COUNT=$((FAIL_COUNT + 1))
+    fi
+}
+
+for vault_name in "${VAULT_NAMES[@]}"; do
+    audit_one "$vault_name"
+done
+
+# --- aggregate (once, after all vaults) ---
+echo ""
+echo "=== aggregate ==="
+"$FN_BIN" vault aggregate 2>&1 | sed 's/^/  /'
+
+# --- doctor (read-only health check) ---
+echo ""
+echo "=== doctor ==="
+"$FN_BIN" vault doctor 2>&1 | sed 's/^/  /' || true
+
+# --- summary table ---
+END_TS=$(date +%s)
+ELAPSED=$(( END_TS - START_TS ))
+
+echo ""
+echo "=== summary ==="
+printf "%-30s  %s\n" "VAULT" "STATUS"
+printf "%-30s  %s\n" "-----" "------"
+for vault_name in "${VAULT_NAMES[@]}"; do
+    status="${VAULT_STATUS[$vault_name]:-unknown}"
+    printf "%-30s  %s\n" "$vault_name" "$status"
+done
+echo ""
+echo "Done: ${PASS_COUNT} ok, ${FAIL_COUNT} warn  (${ELAPSED}s)"
+
+if [[ $FAIL_COUNT -gt 0 ]]; then
+    exit 4
+fi
+exit 0