feat(1c): derive company activity from file-base telemetry

clickhouse-1c/ai/company_intelligence_api.py

@@ -146,8 +146,8 @@ def company_summary(counterparty: str, infobase: str | None = None) -> dict[str,
     signals = rows_to_dict(client.query(signals_sql))
     timeline = rows_to_dict(client.query(timeline_sql))
     essence = (
-        f"Компания {counterparty}: за 30 дней документов {card['docs_30d']}, объём {card['amount_30d']}, "
-        f"прогноз на 30 дней {card['amount_forecast_30d']}, риск {card['signal_severity']}."
+        f"Компания {counterparty}: за 30 дней событий {card['docs_30d']}, суммарная активность {card['amount_30d']}, "
+        f"прогноз активности на 30 дней {card['amount_forecast_30d']}, риск {card['signal_severity']}."
     )
     return {
         "essence": essence,

clickhouse-1c/ai/refresh_company_intelligence.py

@@ -218,7 +218,7 @@ def main() -> int:
         if days_since_last_activity >= 14 and (docs_prev_7d > 0 or amount_prev_7d > 0):
             signals.append(("inactive_company", 85, "high", f"Нет активности по компании {counterparty} уже {days_since_last_activity} дн."))
         if amount_prev_7d > 0 and amount_7d < amount_prev_7d * 0.5:
-            signals.append(("amount_drop", 70, "high", f"Объём по компании {counterparty} упал более чем на 50% неделя к неделе."))
+            signals.append(("amount_drop", 70, "high", f"Активность по компании {counterparty} упала более чем на 50% неделя к неделе."))
         if docs_prev_7d > 0 and docs_7d == 0:
             signals.append(("docs_stopped", 55, "medium", f"По компании {counterparty} прекратился поток документов за последние 7 дней."))
         if open_cases_total > 0:

clickhouse-1c/detections/build_entity_timeline.sql

@@ -11,12 +11,32 @@ FROM (
         concat('document:', doc_type) AS event_type,
         if(posted = 1, 'low', 'medium') AS severity,
         if(posted = 1, 5, 20) AS score,
-        doc_id AS ref_id,
+        concat(doc_id, ':', toString(toUnixTimestamp(ts)), ':', doc_type) AS ref_id,
         concat('Документ ', doc_type, ' №', doc_number, ' статус=', status) AS summary
     FROM analytics_1c.documents
 ) AS src
 WHERE src.ref_id NOT IN (SELECT ref_id FROM analytics_1c.entity_timeline);
 
+INSERT INTO analytics_1c.entity_timeline
+SELECT *
+FROM (
+    SELECT
+        ts,
+        'counterparty' AS entity_type,
+        counterparty AS entity_id,
+        infobase,
+        author AS actor,
+        'documents' AS source,
+        concat('counterparty:', operation_type) AS event_type,
+        if(status = 'busy', 'medium', 'low') AS severity,
+        greatest(10, toUInt32(round(amount))) AS score,
+        concat('counterparty:', counterparty, ':', doc_id, ':', toString(toUnixTimestamp(ts))) AS ref_id,
+        concat('Активность компании ', counterparty, ': ', doc_type, ' score=', toString(amount), ' status=', status) AS summary
+    FROM analytics_1c.documents
+    WHERE counterparty != ''
+) AS src
+WHERE src.ref_id NOT IN (SELECT ref_id FROM analytics_1c.entity_timeline);
+
 INSERT INTO analytics_1c.entity_timeline
 SELECT *
 FROM (

clickhouse-1c/detections/insert_detections.sql

@@ -19,6 +19,26 @@ FROM (
 ) AS src
 WHERE src.detection_id NOT IN (SELECT detection_id FROM analytics_1c.detections);
 
+INSERT INTO analytics_1c.detections
+SELECT *
+FROM (
+    SELECT
+        generated_at AS ts,
+        concat('company_signal:', signal_id, ':', toString(toUnixTimestamp(generated_at))) AS detection_id,
+        infobase,
+        concat('company_', signal_type) AS rule_id,
+        'Сигнал активности компании' AS rule_title,
+        'counterparty' AS entity_type,
+        counterparty AS entity_id,
+        severity,
+        score,
+        summary,
+        'open' AS status
+    FROM analytics_1c.v_company_health_current
+    WHERE severity IN ('medium', 'high', 'critical')
+) AS src
+WHERE src.detection_id NOT IN (SELECT detection_id FROM analytics_1c.detections);
+
 INSERT INTO analytics_1c.detections
 SELECT *
 FROM (

clickhouse-1c/grafana/provisioning/dashboards/files/1c-company-intelligence.json

@@ -79,7 +79,7 @@
     },
     {
       "id": 2,
-      "title": "Прогноз объёма 30д",
+      "title": "Прогноз активности 30д",
       "type": "stat",
       "datasource": {
         "type": "grafana-clickhouse-datasource",
@@ -235,7 +235,7 @@
     },
     {
       "id": 5,
-      "title": "Объём по компаниям",
+      "title": "Активность по компаниям",
       "type": "timeseries",
       "datasource": {
         "type": "grafana-clickhouse-datasource",
@@ -282,7 +282,7 @@
     },
     {
       "id": 6,
-      "title": "Количество документов по компаниям",
+      "title": "Количество событий по компаниям",
       "type": "timeseries",
       "datasource": {
         "type": "grafana-clickhouse-datasource",

clickhouse-1c/ops/run_ingest_cycle.sh

@@ -45,6 +45,14 @@ docker exec -i "${CH_CONTAINER}" clickhouse-client \
   --database "${CLICKHOUSE_DB}" \
   < "${ROOT}/detections/build_entity_timeline.sql"
 
+docker exec -i "${CH_CONTAINER}" clickhouse-client \
+  --user "${CLICKHOUSE_USER}" \
+  --password "${CLICKHOUSE_PASSWORD}" \
+  --database "${CLICKHOUSE_DB}" \
+  < "${ROOT}/clickhouse/init/04_company_intelligence.sql"
+
+"${ROOT}/ops/run_company_intelligence_refresh.sh"
+
 docker exec -i "${CH_CONTAINER}" clickhouse-client \
   --user "${CLICKHOUSE_USER}" \
   --password "${CLICKHOUSE_PASSWORD}" \
@@ -56,11 +64,3 @@ docker exec -i "${CH_CONTAINER}" clickhouse-client \
   --password "${CLICKHOUSE_PASSWORD}" \
   --database "${CLICKHOUSE_DB}" \
   < "${ROOT}/detections/open_cases_from_detections.sql"
-
-docker exec -i "${CH_CONTAINER}" clickhouse-client \
-  --user "${CLICKHOUSE_USER}" \
-  --password "${CLICKHOUSE_PASSWORD}" \
-  --database "${CLICKHOUSE_DB}" \
-  < "${ROOT}/clickhouse/init/04_company_intelligence.sql"
-
-"${ROOT}/ops/run_company_intelligence_refresh.sh"